Abstract: A connected device at a client network implements a local data classification service for classifying data based on a data classification service of a remote provider network. The local data classification service receives a request to classify data at one or more data sources of the client network. The request is initiated from a client device of the client network according to a management interface for a data classification service of a remote provider network (e.g., using the same API request used by the remote classification service). The local data classification service obtains at least some of the data from the one or more data sources of the client network. The local data classification service classifies the obtained data according to different types of sensitivity using the data classification engine in the execution environment without the data being exposed outside of a data isolation boundary of the client network.

Abstract: Methods, systems, and computer storage media for providing a unified multilayer-based index for a contextual geoanalytics engine in a data analytics system. The contextual geoanalytics engine is configured to aggregate point-of-interest geographical data from multiple data sources into an aggregate or composite dataset. The contextual geoanalytics engine then transforms and maps the data into a homogenous dataset—i.e., a location embedding record that is homogenous representation of an aggregated dataset—comparable across global geographical regions. The homogenous dataset is accessible via the unified multilayer-based index that is a single geographical index, where the homogenous dataset is a composite of different datasets. The data includes different data types, where the data types are stored in different layers while sharing a common index (i.e., unified multilayer-based index).

Abstract: A system and method for technology analysis utilizing high-performance, scalable, multitenant, dynamically specifiable, knowledge graph information storage and utilization. The system uses an in-memory associative array for high-performance graph storage and access, with a non-volatile distributed database for scalable backup storage, a scalable, distributed graph service for graph creation, an indexing search engine to increase searching performance, and a graph crawler for graph traversal. One or more of these components may be in the form of a cloud-based service, and in some embodiments the cloud-based services may be containerized to allow for multitenant co-existence with no possibility of data leakage or cross-over. The system uses a cyber-physical graph to represent an enterprise's cyber-physical system and can provide graph analysis, graph security, and graph fusion related tasks to identify potential operational risks.

Abstract: Systems and methods including one or more processors and one or more non-transitory computer readable media storing computing instructions that, when executed on the one or more processors, perform: receiving historical in-session user activity information; receiving, via a graphical user interface (GUI) of a user device, a partial search query from a user; analyzing the partial search query based on the historical in-session user activity information using one or more query suggestion systems to determine a respective score for respective suggested search queries from each of the one or more query suggestion systems; determining a respective absolute position metric for the respective suggested search queries from each of the one or more query suggestion systems; determining a respective efficiency metric for each of the one or more query suggestion systems based on the respective absolute position metric; analyzing the respective efficiency metric for each of the one or more query suggestion systems to dete

Abstract: Systems and methods are provided for creating and utilizing search indexes with different properties to provide efficient and flexible means for retrieving and ranking high-dimensional content. A first index and a second index are generated with a different sets of properties, which may include a permutation prefix length and a vector quantization scheme, and are each associated with a computational workload. Further, at least one index structure may be created with a posting list associated with a permutation prefix and a set of references, rather than only a permutation prefix, to balance out posting lists of variable lengths within the index. Another embodiment of the disclosure provides for expansion of search results at search time, rather than during indexing, by retrieving posting lists associated with shorter permutation prefix after an initial list is determined to have insufficient recall.

Abstract: Systems and methods for personal data classification, linkage and purpose of processing prediction are provided. The system for personal data classification includes an entity extraction module for extracting personal data from one or more data repositories in a computer network or cloud infrastructure, a linkage module coupled to the entity extraction module, a linkage module coupled to the entity extraction module and a processing prediction module. The entity extraction module performs entity recognition from the structured, semi-structured and unstructured records in the one or more data repositories. The linkage module uses graph-based methodology to link the personal data to one or more individuals.

Abstract: A system for executing a streaming query includes an interface and a processor. The interface is configured to receive a logical query plan. The processor is configured to determine a physical query plan based at least in part on the logical query plan. The physical query plan comprises an ordered set of operators. Each operator of the ordered set of operators comprises an operator input mode and an operator output mode. The processor is further configured to execute the physical query plan using the operator input mode and the operator output mode for each operator of the query.

Abstract: According to an embodiment, a display control apparatus includes a clustering unit, a sub-clustering unit, and a display control unit. The clustering unit is configured to classify images into a plurality of clusters based on similarity degrees of the images and a first threshold. The sub-clustering unit is configured to further classify images within each of the plurality of clusters into a plurality of sub-clusters based on the similarity degrees and a second threshold that is higher than the first threshold. The display control unit is configured to display, on a display unit, display information including a cluster representative indicative of a representative of images included in a cluster and a sub-cluster representative indicative of a representative of an image included in a sub-cluster.

Abstract: Adaptive matching is provided, which is used to automatically match a block size of a transactional file system with an IO size of a client. An example method includes creating a share domain, wherein the share domain is created on a first file system, and a block size of the first file system is a first block size. The method further includes determining a block size of the share domain as a second block size, wherein the second block size is not equal to the first block size. If a block size of a second file system is the second block size, the share domain is migrated from the first file system to the second file system. By implementing the present application, it is possible to simplify user operations, improve operational convenience, and help to reduce storage space fragments and indirect blocks, thereby further improving the performance of a storage system.

Abstract: A linguistic system includes a controller. The controller may be configured to receive a query and document, tokenize the query into a sequence of query tokens and tokenize the document into a sequence of document tokens, generate a matrix of token pairs for each of the query and the document tokens, retrieve for each entry in the matrix of token pairs, a precomputed similarity score produced by a neural conditional translation probability network, wherein the neural network has been trained in a ranking task using a corpus of paired queries and respective relevant documents, produce a ranking score for each document with respect to each query via a product-of-sum aggregation of each of the similarity scores for the respective query; and output the document and associated ranking score of the document.

Abstract: Keywords obtained from a user and/or extracted from uploaded document(s) may be used to generate potential keywords. Documents may be identified based on the keywords and the potential keywords accepted by the user. A knowledge graph model representing the identified documents may be generated. The knowledge graph model may include document nodes representing the identified document and a search node representing the keywords. The relative position of the document nodes with respect to the search node may represent similarity between the corresponding documents and the keywords.

Abstract: The present disclosure relates to search techniques. In one example process, the device concurrently displays remote search results and local search results. In another example process, the device provides previews of search results that include actionable user interface objects. In another example process, the device concurrently displays options for initiating a search using various search engines.

Abstract: An apparatus comprises a processing device configured to select a first data field of a first type that is associated with a second data field of a second type in a document, to determine an embedding of terms of unstructured text data in the first data field and to identify a subset of paired data fields from an unstructured text database based at least in part on metrics characterizing similarity between (i) the embedding of terms in the first data field and (ii) embeddings of terms in data fields of the first type in the paired data fields. The processing device is further configured to determine syntactic differences between the unstructured text data in the first data field and the identified subset of paired data fields, and to provide recommendations for unstructured text data to fill the second data field in the document based on the syntactic differences.

Abstract: As digital assets are created and shared, design components that comprise those digital assets are curated, organized, and tracked so as to allow meaningful relationships to be established between shared assets and design components. The tracking that underlies such relationships is provided by metadata associated with a given design component. This metadata may include information such as an asset identifier that identifies a source digital asset from which the design component was extracted; a version identifier that identifies a version of the source digital asset; an author identifier that identifies an author of the source digital asset; and a layer identifier that can be used to reveal the context in which the design component was derived from the source asset. This metadata allows relationships to be established between a design component and the digital assets that incorporate that design component, thus facilitating asset and component tracking and update notification broadcasting.

Abstract: A method, apparatus, system, and computer program code for performing a human resource operation using a context-independent conversational flow. The computer system receives an intended human resource operation from an application executing on a user device, identifies the context-independent conversational flow for performing an intended human resource operation, and calls a structured data object according to the context-independent conversational flow. The computer system interprets the structured data object to produce a business rule output, and generates a context-independent response from the business rule output. The computer system transforms the context-independent response according to a user context to produce a context-specific response, and forwards the context-specific response to a user device for display within a conversational user interface of an application.

Abstract: Digital content search techniques are described. In one example, the techniques are incorporated as part of a multi-head self-attention module of a transformer using machine learning. A localized self-attention module, for instance, is incorporated as part of the multi-head self-attention module that applies local constraints to the sequence. This is performable in a variety of ways. In a first instance, a model-based local encoder is used, examples of which include a fixed-depth recurrent neural network (RNN) and a convolutional network. In a second instance, a masking-based local encoder is used, examples of which include use of a fixed window, Gaussian initialization, and an adaptive predictor.

Abstract: A file moving method, an electronic device, and a computer-readable storage medium are provided. An implementation includes: in response to receiving a file moving request, determining a source path and a destination path corresponding to a first moving operation corresponding to the request; determining whether the source path and the destination path corresponding to the first moving operation exist; in response to determining that both the source path and the destination path corresponding to the first moving operation exist, comparing at least one of the source path and the destination path corresponding to the first moving operation with a corresponding path of a source path and a destination path corresponding to the second moving operation; and determining, based on a result of the comparing, whether the first moving operation is to be performed.

Abstract: Techniques are described herein for training and/or utilizing a query canonicalization system. In various implementations, a query canonicalization system can include a classification model and a canonicalization model. A classification model can be used to determine if a search query is well-formed. Additionally or alternatively, a canonicalization model can be used to determine a well-formed variant of a search query in response to determining a search query is not well-formed. In various implementations, a canonicalization model portion of a query canonicalization system can be a sequence to sequence model.

Abstract: A metadata query is received in as storage system from a client node that asks one or more questions regarding a set of metadata in a metadata store. The metadata query is executed on at least one metadata summary structure. Each metadata summary structure summarizes a metadata stream of metadata access during a period of time, and different metadata summary structures from different periods of time are periodically merged together so that the metadata query is conducted on a merged metadata summary structure. An answer is returned from the query executed on the merged metadata summary structure.

Abstract: Systems and methods for quantifying saved time during data entry.