An inverted index is a lookup table that allows a user to quickly find all of the records that have a common value. For example, all of the records that the value "JOHN" for the first name. The principle of the inverted index can be applied to entity resolution (ER) algorithms as technique for reducing the number of comparisons needed to resolve a set of input references.
In the case of ER, the common value is a match-key. When an input reference is read into the OYSTER engine for processing, the index quickly locates all previously processed records that share the same match key as the input reference.

Back to OYSTER Reference Guide page

Click Prev 7-OYSTER User Defined Inverted Index page

Click Next Match Key page