The disclosure discloses a method for analyzing educational big data on the basis of maps. The method includes acquiring educational resource data and storing the educational resource data into databases according to certain data structures; constructing theme map layers for each analysis theme, classifying and indexing data according to the analysis themes, and superimposing the theme map layers onto base maps to form data maps; analyzing data of the theme map layers according to the analysis themes and acquiring theme analysis results; extracting the data of the multiple theme map layers in target regions, fusing the data and acquiring region analysis results; acquiring learning preference of users; combining the learning preference of the users according to content of user requests and searching the region analysis results in response to the user requests. The disclosure further discloses a system for analyzing the educational big data on the basis of the maps.

This disclosure relates to methods and systems for processing of sequencing data streams. The system receives sequences from a sequencer and stores them as data records on a database. The sequences are associated with a counter indicative of a number of times the associated sequence has been sequenced. The system progressively receives a further sequence as streaming data from the sequence. While receiving the further sequence, the system matches the streaming data against the stored sequences to determine a matching score. Upon the matching score exceeding a matching threshold for one of the multiple sequences in the database, the system selects the one of the sequences in the database based on the matching score and stores the further sequence on non-volatile memory where the counter value associated with the selected sequence is below a saturation threshold. The system also terminates the receiving where the counter value is above the saturation threshold.

Embodiments of the invention utilize a “data canopy” that breaks statistical measures down to basic primitives for various data portions and stores the basic aggregates in a library within an in-memory data structure. When a queried statistical measure involves a basic aggregate stored in the library over a data portion that at least partially overlaps the data portion associated with the basic aggregate, the basic aggregate may be reused in the statistical computation of the queried measure.

Systems, methods, and computer-readable media for determining at least one valid mood for a collection of media content items of a media library are provided.

A method and system for providing on-demand updating of metadata information associated with clinical image data files is disclosed. The system allows for extracting, on the fly, additional metadata fields from clinical image data files and re-indexing the metadata to include the additional fields so that an end-user can perform operations (e.g., searching, browsing, etc.) based on the newly indexed fields. The metadata fields may be defined in a configurable schema file (e.g., XML, YML, etc.). The system can gain efficiencies by, for example, reading only a subset of the DICOM file (e.g., reading only the first few kilobytes that contain the header), scanning only a subset of source data folders, etc.

The present disclosure relates to concurrent learning of a relationship estimation model and a phrase generation model. The relationship estimation model estimates a relationship between phrases. The phrase generation model generates a phrase that relates to an input phrase. The phrase generation model includes an encoder and a decoder. The encoder converts a phrase into a vector using a three-piece set as learning data. The decoder generates, based on the converted vector and a connection expression or a relationship label, a phrase having a relationship expressed by the connection expression or the relationship label for the phrase. The relationship estimation model generates a relationship score from the converted vector, which indicates each phrase included in a combination of the phrases, and a vector indicating the connection expression and the relationship label.

One aspect of the subject matter described herein comprises a database management system. The database management system comprises a communication circuit, an interface, and a processor. The communication circuit receives information from databases via a communication network. The interface allows user operation and interaction via the communication network and the database management system. The processor provides data to and receives data from the interface, including a plurality of alphanumeric records comprising at least one unique identifier, obtains database records from other databases for each unique identifier, identifies a number of alphanumeric records in the plurality having events in their histories, determines a percentage of alphanumeric records in the plurality having negative events in their histories, generates a report including the determined percentages, and conveys the report to the user.

A computer system is provided that is programmed to select feature sets from a large number of features. Features for a set are selected based on metagradient information returned from a machine learning process that has been performed on an earlier selected feature set. The process can iterate until a selected feature set converges or otherwise meets or exceeds a given threshold.

An electronic apparatus for compressing a recurrent neural network and a method thereof are provided. The electronic apparatus and the method thereof include a sparsification technique for the recurrent neural network, obtaining first to third multiplicative variables to learn the recurrent neural network, and performing sparsification for the recurrent neural network to compress the recurrent neural network.

Provided is a database system and method in which storage is partitioned in an unbalanced format for faster access. In one example, the method may include one or more of receiving a request to store a data record, identifying a partition from among a plurality of partitions of a database based on a shard identifier in the request, automatically determining a unique range of data identifiers designated to the partition from the plurality of partitions, respectively, based on an unbalanced partitioning, determining whether the data identifier is available within the unique range of data identifiers of the identified partition, and storing the data record at the identified partition in response to determining the data identifier is available. The unbalanced partitioning according to various embodiments reduces the partitions that need to be checked during a data insert/access operation of the database.