34results about "Semi-structured data retrieval" patented technology

In order to apply a text-based label inference approach to infer a label to be assigned to target data in a common data form, an inference apparatus (1) includes: a data converting section (11) for converting target data subject to label assignment into text; and a label inferring section (12) for inferring a label to be assigned to the target data, in accordance with a label inference model for inferring a label to be assigned to text and the text obtained by the converting carried out by the data converting means.

The invention discloses a tracking method of a semi-structured dialogue state, a medium and electronic equipment, and belongs to the technical field of computers, a dialogue state model based on UML is designed to represent and track the dialogue state, the dialogue state model comprises multiple rounds of decision making processes, priori knowledge of a service is abstracted, and the decision making efficiency is improved. According to the method, the information density of a single dialogue sample is increased in the state of each round of dialogue process, so that the search space of the dialogue state is reduced, the requirement of a DST model for the number of training corpus samples is reduced, the cold start problem of model training is solved, the samples are generated on the basis of the dialogue state model, and the efficiency is improved. The DST model can continuously learn and adjust in a dynamic environment within a certain range through reinforcement learning to adapt to new data or tasks, the content output by the model can be controllable, the precision of processing complex tasks is improved, and the problem that the model cannot track a long dialogue state is solved.

The technology provides a rich attention mechanism for structured information extraction of web pages and other electronic documents. An input layer of a model obtains system, information associated with the document, including field tokens representing respective fields to be extracted from the document, structured document type tokens associated, and text tokens from a text sequence in the document. An encoder connects the field tokens, the S type tokens and the text tokens according to a set of different attention patterns. The encoder generates an overall token representation based on the set of different attention patterns. An output layer of the model extracts a final text span for the each of the respective fields from the set of text tokens. The extracted final text span for each of the respective fields is stored in memory, and can be produced in response to a search query, analytics evaluation or other request.

The invention discloses an electric power system XML generation method and system based on mxGraph and G6, and relates to the technical field of computer platform load balancing, and the method comprises the steps: collecting first object parameter data, analyzing the parameter data to obtain influence parameters, converting the influence parameters to obtain graph model data, and storing the graph model data in a database; and establishing a format conversion model to analyze and adjust the graph model data to obtain XML format data. According to the power system XML generation method based on mxGraph and G6, through combination of mxGraph and G6 technologies, the system can provide more flexible graphical operation and real-time feedback, and the interactive experience of a user in the power system design process is enhanced. An efficient graph model data analysis and conversion mechanism is introduced, so that the power system data can be quickly and accurately converted into a standardized XML format.

An embodiment of the present invention discloses a unified description and shared discovery method for intelligent models, comprising the following steps: S10: using PMML to describe the relevant information of the intelligent model to form a PMML model description document, storing the PMML model description document in an XML database, and storing the intelligent model in an intelligent model library; S30: encapsulating the intelligent model to form an intelligent simulation model; S50: responding to an execution instruction to realize the shared discovery of the intelligent model.

The present disclosure relates, inter alia, to a computer-implemented read-only data structure for archiving snapshots of an RDF dataset into a file, the RDF dataset comprising one or more graphs, the read-only data structure being able to be queried directly using all possible triple patterns of a SPARQL query engine. The data structure includes: an indexed list of one or more graphs of an RDF data set; for each graph of the RDF dataset, an indexed list of predicates; mapping each RDF lexical item of the data set to a corresponding index and a dictionary mapped in reverse; and for each predicate indexed in the indexed list of predicates of each graph, representing an adjacency matrix of a set of tuples of the RDF dataset, thereby obtaining a complete representation of the RDF dataset that can be queried directly using all possible triple patterns of the SPARQL query engine.

A system and method of the disclosure relates to structuring at least one resource of an application programming interface (API) for a server. In the method, a plurality of field-level elements may be defined. Each of the plurality of field-level elements may be associated with a respective property. Also, first and second message-level elements may be defined. The first message-level element may be based on a first field-level element of the plurality of field-level elements, and the second message-level element may be based on the first field-level element. The API may be associated with a first resource having properties resolved based on at least the first and second message-level elements.

A computer-implemented method for generating by a query engine a graph of operators for a SPARQL query over an RDF graph. The method includes obtaining a graph of operators executable by the query engine, the graph comprising a plurality of basic operators, at least two of said operators being of a first type each configured to find RDF triples of the RDF graph that match a respective basic graph pattern. The method further comprises identifying a group of operators among the at least two basic operators of the graph which are of the first type. The respective basic graph patterns of the group of operators have same subject and / or predicate and / or object and the identified group of operators is replaced in the graph by an equivalent operator configured to find RDF triples of the RDF graph that match the respective basic graph patterns of the group of operators.

Data loss prevention may be applied to data that includes a plurality of records, with each record including a plurality of fields and with each field corresponding to a different one of a plurality of categories. Applying of data loss prevention may include selecting a subset of records, computing, for each category, a likelihood the category contains the sensitive information, selecting a subset of categories based on the computed likelihoods, searching the sensitive information in the selected subset of categories, and in response to detection of sensitive information in at least one of the subset of records, taking one or more data loss prevention related actions. The selected subset of records includes fewer records than the plurality of records, may include a number of non-consecutive records, and the subset of records may be selected such that the records are well distributed within the plurality of records of the data.

The present invention relates to context-aware feature embedding using deep recursive neural networks and anomaly detection of sequential log data. Techniques are provided herein for contextually embedding features of network traffic or operation logs for anomaly detection based on sequential prediction. In an embodiment, a computer has a predictive recursive neural network (RNN) for detecting anomalous network flows. In an embodiment, the RNN transcodes a sparse feature vector representing a log message into a dense feature vector in context, which may be predictive or used to generate a predictive vector. In an embodiment, graph embedding improves the feature embedding of log traces. In an embodiment, the computer detects and feature-encodes independent traces from related log messages. These techniques can detect malicious activities by performing anomaly analysis on context-aware feature embedding of network packet flows, log messages, and / or log traces.

To solve a problem in which a user of data created by a creator may have difficulty in extracting necessary information from the data, the creator being allowed to freely select a format of the spreadsheet data.SOLUTION: A data processing system 10 includes a first processor 22, a first storage device 24, a second input device 36, and a second output device 38. The first processor 22 is configured to: (a) acquire an input file including tabular data to be stored in a first storage device 24; (b) output an interface to the second output device 38, for inputting first information that defines element relationships between a plurality of elements in the tabular data; (c) generate semi-structured data including the element relationships defined by the input first information from the tabular data included in the input file stored in the first storage device 24; and (d) generate an output file including the generated semi-structured data.SELECTED DRAWING: Figure 10

The application provides a lock execution method and device based on a transactional key-value database, equipment and medium, which can be applied to the field of electric digital data processing. The method comprises the following steps: in response to a first transaction operation request for a target resource in a transactional key-value database, reading a first lock key and a first lock value of a distributed lock corresponding to the target resource, wherein the first lock value comprises an instance identifier holding the distributed lock and valid timestamp information of the distributed lock, and the first transaction operation request comprises a target instance identifier and a lease duration of the distributed lock; determining an occupation result of the distributed lock according to the first lock key and the first lock value; in the case that the occupation result represents that the distributed lock is not occupied, updating the instance identifier and the valid timestamp information in the first lock value according to the target instance identifier and the lease duration, so that the target instance executes a transaction operation indicated by the first transaction operation request.

The invention provides a lock execution method, device and equipment based on a transactional key value database and a medium, which can be applied to the field of electric digital data processing. The method comprises the steps that in response to a first transaction operation request for a target resource in a transactional key value database, a first lock key and a first lock value of a distributed lock corresponding to the target resource are read, the first lock value comprises an instance identifier with the distributed lock and effective timestamp information of the distributed lock, and the instance identifier and the effective timestamp information of the distributed lock correspond to the target resource; the first transaction operation request comprises a target instance identifier and lease duration of the distributed lock; determining an occupation result of the distributed lock according to the first lock key and the first lock value; and under the condition that the occupation result represents that the distributed lock is not occupied, according to the target instance identifier and the lease duration, updating the instance identifier and the effective timestamp information in the first lock value, so that the target instance executes a transaction operation indicated by the first transaction operation request.

Systems and Methods for Presenting Search Results and Computer Readable Media. A system identifies a document related to a search term, in which the document includes a set of structural elements. The system determines a distribution of occurrences of the search term in the document, identifies one of the structural elements, based on the distribution of occurrences of the search term in the document, and presents information associated with the identified structural element.

This real-time data processing system, based on Web architecture Canvas technology, belongs to the fields of automated report logic engine technology and web page table processing technology. The invention includes a data extraction module, a data parsing module, a data storage module, a real-time display module, and an online modification module. The data extraction module dynamically extracts business data from the business system to be analyzed. The data parsing module parses the extracted business data using a built-in parsing model and a custom parsing model. The data storage module stores the business data and the results of the data parsing. The online modification module performs real-time online modification and statistical analysis on the data to be displayed, while synchronizing the modified data to the data storage module. The real-time display module provides an interface with the operator. This invention is compatible with historical data from traditional business personnel and allows business personnel to manually configure and display data results in real time.

Method for data loss prevention in data, wherein the data are structured and / or semi-structured data with a plurality of records and a plurality of categories, wherein each record comprises a plurality of fields each belonging to a different one of the categories, the method comprising the following steps: selecting, in a processor, a subset of records of the data, wherein the selected subset comprises at least two records between which the data comprise a plurality of records which are not included in the subset; scanning, in the processor, the fields of the selected subset of categories for sensitive information; computing, in the processor, for each category a likelihood to contain the sensitive information on the basis of the scanning result; selecting, in the processor, a subset of categories on the basis of the computed likelihoods of the categories to contain the sensitive information; and searching, in the processor, the sensitive information in the selected subset of categories.

Technologies are described herein for executing queries expressed with reference to a structured query language against unstructured data. A user issues a structured query through a traditional structured data management (“SDM”) application. Upon receiving the structured query, an SDM driver analyzes the structured query and extracts a data structure from the unstructured data, if necessary. The structured query is then converted to an unstructured query based on the extracted data structure. The converted unstructured query may then be executed against the unstructured data. Results from the query are reorganized into structured data utilizing the extracted data structure and are then presented to the user through the SDM application.

The application discloses a page embedding method and related equipment. The method comprises the following steps: obtaining a link address of a first page description file and a compiling script corresponding to the first page description file, the compiling script being used for parsing the first page description file, and the first page description file being used for describing a first page; embedding the compiling script in a second page description file, the second page description file being used for describing a second page, and the domain name of the second page being different from that of the first page; obtaining the first page description file based on the link address through the compiling script, embedding description information of the first page description file on the first page into the second page description file to obtain a third page description file, and the third page description file being used for describing the second page after the first page is embedded; and loading the third page description file.

The invention relates to the technical field of data processing, in particular to a data analysis method and system based on a large model, and the system comprises a knowledge base management module which is used for creating and managing a knowledge base; the visualization construction module is used for arranging data analysis tasks in a visualization mode based on business requirements; the data analysis task management module is used for processing the data analysis task based on the knowledge base to obtain a processing result; the task debugging module is used for debugging the data analysis task so as to track the processing flow of the data analysis task and track whether the processing result meets a preset condition or not; and the task output module is used for outputting the data analysis task of which the processing result meets the preset condition through a preset service interface. According to the system, efficient and intelligent data analysis is realized, and full-process visual management of a data analysis task is realized.

The present invention relates to common agreed domain models, data models, and protocols for providing open-to-box interoperability between different systems (e.g., including systems from different manufacturers) in a shared context defined based on "Knowledge Requirements". Therefore, more efficient integration between products is realized.

In one embodiment, a method for creating one or more infographics, comprising: receiving and storing data associated with an individual or an entity, in a format according to a schema that includes at least two properties associated with the individual or entity; reading at least a portion of the data; determining which of the at least two properties in the schema do not have corresponding read data associated with the individual or entity; based on that determination, selecting an infographic definition from among a plurality of infographic definitions defining the appearance of at least a portion of an infographic; generating one or more infographics based on (i) the at least a portion of the data and (ii) the selected infographic definition; and providing the one or more generated infographics to an output device.

The application relates to a graph database adjacency list cache management method based on dynamic weights, which comprises the following steps: receiving a request for querying a node adjacency list, generating a new cache item according to the adjacency list of the node if the request is not hit, determining the inherent weight of the cache item based on the inherent attributes of the node, setting the heat value of the cache item as an initial heat value, and obtaining the comprehensive weight of the cache item based on the inherent weight and the heat value; if the request is hit, obtaining the historical heat value of the cache item, determining a new heat value based on a smooth increment strategy and the historical heat value, performing weighted summation on the new heat value and the inherent weight of the cache item to obtain the comprehensive weight of the cache item, and deleting the cache item with the minimum comprehensive weight under the condition that the cache space is insufficient. Through the application, the problem of low cache hit rate of a graph database is solved, the inherent weight avoids the adjacency list of a super node from being occupied by common nodes, the heat value can take into account the common nodes that are suddenly popular, and the overall cache hit rate is improved.

The disclosure notably relates to a computer-implemented read-only data structure for archiving a snapshot of a RDF dataset into a file, the RDF dataset comprising one or more graphs, the read-only data structure that can be directly queried using all possible triple patterns of a SPARQL query engine. The data structure comprises an indexed list of the one or more graphs of the RDF dataset, for each graph of the RDF dataset, an indexed list of predicates, a dictionary mapping each RDF term of the dataset to a respective index and inversely, and for each predicate indexed in the indexed list of predicates of each graph, an adjacency matrix representing a group of tuples of the RDF dataset, thereby obtaining a complete representation of the RDF dataset that can be directly queried using all possible triple patterns of a SPARQL query engine.

A computer-implemented read-only data structure for archiving a snapshot of a RDF dataset into a file, the RDF dataset comprising one or more graphs, the read-only data structure that can be directly queried using all possible triple patterns of a SPARQL query engine. The data structure comprises an indexed list of the one or more graphs of the RDF dataset, for each graph of the RDF dataset, an indexed list of predicates, a dictionary mapping each RDF term of the dataset to a respective index and inversely, and for each predicate indexed in the indexed list of predicates of each graph, an adjacency matrix representing a group of tuples of the RDF dataset, thereby obtaining a complete representation of the RDF dataset that can be directly queried using all possible triple patterns of a SPARQL query engine.

In order to estimate a label to be given to target data of a general data format, by applying a text-based label estimation method, this estimation device (1) comprises: a data conversion unit (11) that converts target data, to which a label is given, into text; and a label estimation unit (12) that estimates a label to be given to the target data, on the basis of a label estimation model for estimating the label to be given to the text, and the text obtained by the conversion by the data conversion unit.

The invention relates to the technical field of automatic intelligent operation inspection of a power system, and provides an intelligent substation maintenance work permission and acceptance section data comparison method and system, and the method comprises the steps: defining a maintenance work acceptance general item set; analyzing the maintenance work acceptance general item set to obtain a specific acceptance sub-item type involved in each item set, and performing digital signal mapping on the acceptance sub-items to obtain an intelligent substation maintenance work acceptance data signal point set; constructing a whole station section state comparison data set model; all intervals or part of intervals of the whole intelligent substation are selected, and work permission section data are frozen into an XML self-description file; when the maintenance work is finished, comparing the work acceptance section state data with the work permission section data; and outputting a comparison result, and carrying out visual display according to data type layering and classification. Effective assistance can be provided for daily operation and inspection work of the intelligent substation, and the automation level and reliability of inspection and acceptance work are improved.

The embodiment of the invention provides a method and a device for performing static SQL (Structured Query Language) reduction analysis on an XML (Extensible Markup Language) file and a storage medium. The method comprises the following steps of: obtaining a to-be-analyzed XML file, analyzing the XML file to obtain a plurality of elements contained in the XML file, and obtaining a plurality of node categories corresponding to the plurality of elements according to a preset mapping relationship between the elements and the node categories; analyzing each node category in the plurality of node categories one by one to obtain sub-nodes under each node category, sequentially constructing syntax trees by the at least one sub-node according to the sequence, and processing the plurality of syntax trees to generate a category graph; traversing and accessing the class diagram to obtain a final SQL, extracting a field mapping relation from the Resu ltMap to obtain a first attribute corresponding to a field, analyzing the final SQL to obtain a second attribute of the SQL field, determining whether the first attribute is matched with the second attribute or not, and if the first attribute is not matched with the second attribute, giving an alarm to the field.

The invention discloses a competition information management system and method, and relates to the technical field of information management, and the system comprises a competition information collection processing unit, a node fragmentation management unit, a node adaptive management unit, an information storage management unit and a management information output unit. The technical problems that node fragmentation and performance management cannot adapt to competition dynamic characteristics, hierarchical management is not performed during data storage, and storage resources are wasted are solved, the capacity expansion response time is shortened through prediction based on an LSTM + attention mechanism model and competition historical cycle data, and the capacity expansion efficiency is improved through competition type-resource template matching. The resource waste rate is reduced; metadata and business data are separately stored; a centralized shared pool and a local cache are combined; the repeated storage of the metadata is reduced; and the storage cost and the hot data read-write delay are reduced.

The application provides an xml file parsing method, system, computer and readable storage medium, the method comprises the following steps: loading the parsing configuration instruction through the parsing engine; receiving the xml file set, and parsing each xml file in the xml file set through the parsing task flow to parse each xml file into the corresponding object respectively; reading the object into the context manager, and calling the corresponding rule calculator through the parsing configuration instruction; calculating the corresponding value of the object through reading the rule to calculate the attribute value corresponding to each xml file respectively; storing the attribute value in the context manager through the preset path, and making the context manager output the complete parsing structured file. Through the above method, the parsing of xml files with any structure can be realized, and the parsing efficiency of xml files is improved.

The invention relates to a graph database adjacency list cache management method based on a dynamic weight, and the method comprises the steps: receiving a request for querying a node adjacency list, if the request is not hit, generating a new cache item according to the node adjacency list, determining the inherent weight of the cache item based on the inherent attribute of the node, and storing the cache item in a database; setting the popularity value of the cache item as an initial popularity value, and obtaining the comprehensive weight of the cache item based on the inherent weight and the popularity value; if the request is hit, obtaining a historical popularity value of the cache item, determining a new popularity value based on a smooth incremental strategy and the historical popularity value, carrying out weighted summation on the new popularity value and the inherent weight of the cache item to obtain a comprehensive weight of the cache item, and deleting the cache item with the minimum comprehensive weight under the condition that the cache space is insufficient. Through the method and the device, the problem of low cache hit rate of the graph database is solved, the inherent weight prevents the adjacency list of the super node from being occupied by the common node, the heat value can consider the suddenly hot common node, and the overall cache hit rate is improved.