Semantic discovery and mapping between data sources
View Patent ↗An apparatus and method are described for the discovery of semantics, relationships and mappings between data in different software applications, databases, files, reports, messages, or systems. In one aspect, semantics and relationships and mappings are identified between a first and a second data source. A binding condition is discovered between portions of data in the first and the second data source. The binding condition is used to discover correlations between portions of data in the first and the second data source. The binding condition and the correlations are used to discover a transformation function between portions of data in the first and the second data source.
1. A method for determining a mapping between a plurality of columns of a first data source and a second data source by using a fuzzy match, the method comprising:
identifying a join condition between the plurality of columns of the first data source and the second data source;
constructing source data objects based on a set of relationships between the plurality of columns of the first data source and the second data source and the identified join condition;
determining a fuzzy match between the plurality of columns of the first data source and the second data;
identifying the mapping corresponding to the fuzzy match between the plurality of columns of the first data source and the second data source; and
generating a window providing a display of success for the mapping between the columns of the first data source and the second data source.
2. The method as recited in claim 1 , wherein the determining the fuzzy match comprises:
matching the columns of the first data source with the second data source;
identifying a score based on degree of matching of the columns of the first data source with the columns of the second data source; and
identifying a binding condition based on the score of matching, when the score of matching is more than a threshold.
3. The method as recited in claim 1 , wherein identifying the join condition between the plurality of columns of the first data source and the second data source, comprises:
discovering key columns, wherein the key columns are based on a unique index of the columns in the first data source and the second data source; and
discovering a foreign key, wherein the foreign key is based on value match between the discovered key columns.
4. A method for determining a mapping between a plurality of columns of a first data source and a second data source by using a fuzzy match, the method comprising:
identifying a join condition between the plurality of columns of the first data source and the second data source;
constructing source data objects based on a set of relationships between the plurality of columns of the first data source and the second data source and the identified join condition;
determining a fuzzy match between the plurality of columns of the first data source and the second data source by performing:
identifying a list of pre-loaded synonyms corresponding to the columns in the first data source;
comparing the list of pre-loaded synonyms to the columns in the second data source;
identifying a relevancy score based on the comparison between the list of the pre-loaded synonyms and the columns in the second data source; and
determining the mapping based on the relevancy score; and
identifying the mapping corresponding to the fuzzy match between the plurality of columns of the first data source and the second data source.
5. The method as recited in claim 4 , wherein the mapping is used to identify a binding condition between the first data source and the second data source.
6. The method as recited in claim 4 , wherein the list of synonyms is discovered.