40 results about "Schema matching" patented technology

A method and system for performing data analysis where multiple, categorical dimensions and statistical measures are used. The method includes preparing multi-dimensional data into nodes and arranging them logically in the form of a partial-order database. The database may then be traversed by a data-mining tool, or queried in an OLAP manner. Queries are answered through a process of associative partial-order search, which includes optional support for “children-of” requests, and optional support for fuzzy pattern-matching. By aggregating measures on-the-fly, this method is able to preform well without requiring a large number of summary nodes, when taken as a percentage of the total node count.

Tools and methods for schema matching that generate schema graphs, populate match matrices and display the schema graphs and the match matrices. These tools and methods characterize potential matches between disparate schemata in terms of both a strength of evidence indicating the potential match and an amount of evidence indicating the potential match. A number of match voters generate a set of match scores for each potential match, and these match scores are combined by a vote merger to form a single confidence value for each potential match. A number of filters display the confidence value for each potential match as a link on a graphical user interface. Machine-learning techniques may be employed to adaptively determine confidence values based on previously established matches.

Methods and apparatus are provided for improved schema mapping of source documents to target documents. A list of matches are generated between at least one source table and at least one target table. One or more of the matches are annotated with a logical condition providing a context in which the match applies. Matches can be annotated with a logical condition, for example, by generating a set of candidate view conditions, C, to be applied to the one or more source tables. A schema match algorithm can generate the list of matches. Candidate logical conditions can be identified, for example, by (i) creating a set of views for categorical attributes in the tables and adding a view for each partitioning of the attribute values; (ii) using a classifier built on target attribute values; or (iii) evaluating internal features of a source table.

Described are system, method, and computer-program product embodiments for automatically ingesting data from disparate data sources into a target database having a target data schema. In some embodiments, the data is received in a data file including data columns formatted according to a data schema, and a data dictionary describing the source data columns. Count data is generated for each cell selected from a data column, each count datum including counts of occurrences of a detected characteristic in each cell. One or more target data columns from the target data schema can be selected and displayed to a user as being semantically related to the data column based on the count data for each cell, a column header of the data column, and the data dictionary. Based on input received from the user, a data table is generated to store the source columns and loaded into the target database.

The invention discloses an online recognition and filling method for the hourly energy consumption abnormality data of an office building, and the method comprises the following steps: obtaining historical energy consumption data; clearing the historical energy consumption data, and eliminating the abnormality data; building an energy consumption mode feature set according to the outdoor temperature, moment and date attributes of the hourly energy consumption data, and calculating a feature value; collecting the energy consumption data hour by hour, carrying out the energy consumption mode matching, and finding a corresponding feature value; carrying out the online recognition of abnormal data; carrying out the cyclic interpolation of the abnormality data, and carrying out the correction of the interpolation value through a formula; storing the data without abnormality into a database after interpolation; carrying out the rolling correction of the feature values of the energy consumption mode regularly, and then carrying out the online interpolation of the hourly energy consumption data. The method can solve a problem of the recognition and interpolation of the usual abnormality data in the hourly energy consumption of the office building, improves the energy consumption data quality, provides effective data for the further data mining, and promotes the development of the energy saving operation of the building.

Systems and methods for performing schema matching of multiple data dictionaries. Lexical graphs are generated for two or more data dictionaries. The lexical graphs include terms with definitions and identified links between terms. A data processing device determines if matches occur between the generated lexical graphs and outputs the result of the determination.

The embodiment of the invention provides a multi-source data deep fusion method based on deep learning. The method comprises the steps of obtaining a to-be-fused relational data table; constructing adeep learning model, importing training data to carry out word vectorization processing on contents in the to-be-fused relational data table, and carrying out mode matching on the processed data; performing stratified sampling on the data in the to-be-fused relational data table based on the similarity between entities corresponding to the data, importing the sampled data into a preset structure model to perform integration processing based on word vectors to obtain a trained data bucket model, and performing entity-based data bucket processing based on the data bucket model; and judging whether the data in each bucket refers to the same entity, and performing data fusion on the data referring to the same entity to obtain a data table formed by the fused data. According to the invention, character string data is modeled in a word vector mode, and the method can be used for modeling texts and semantics of the character string at the same time, so that the tolerance to dirty data is improved.

The subject disclosure relates to third party cloud schema matching to a user's data. The user can select the type of data to be looked up from an external source in order for the user's data to be augmented with data from the external source. The user can be presented with a list of service providers that support the selected type of data. Columns from the user's data are associated with supported lookup information types either automatically or manually. Based on the association, data from the external data source is automatically mapped to the user's data and updates are selectively applied to the user's data.

The invention provides a geological ontology-based geological report text information extraction method, which comprises the steps of S1, performing preprocessing operation on a geological report document, converting a file type into a data source format, and performing sentence segmentation, word segmentation, stop word removal and part-of-speech tagging; S2, constructing a place name dictionarylibrary and a geological entity dictionary library by utilizing the structured information, expanding on the basis of the existing geological domain ontology, and forming a place name ontology and a geological time ontology; and S3, extracting geological entity information, space-time relationship information and attribute information from the geological report text through a mode matching methodand a rule matching method. The method has the advantages that a large number of training data sets do not need to be marked manually, and manpower and material resources are saved; an existing information extraction model is abstracted and improved, and a certain thought is provided for information extraction in other fields.

The invention discloses an enterprise subject matching method based on part-of-speech tagging of government affair text data. The method comprises the steps: extracting enterprise names in the government affair text data, extracting enterprise naming modes according to known enterprise naming rules, matching the enterprise naming modes, According to the matching result of the enterprise naming modes, carrying out part-of-speech tagging of the government affair text data, carrying out part-of-speech tagging of the government affair text data. according to the enterprise main body matching method, when the to-be-matched government affair data is processed, matching of the enterprise entities can be directly achieved from the government affair text data, the matching efficiency is greatly improved, and intelligent intercommunication of the government affair big data is achieved.