Lucene Search

Relativity allows users to use Lucene Search in on long text fields stored in Data Grid. So if your workspace is enabled for Data Grid, in the Search panel you click Add Condition, and then select Index Search. When the separate Index Search window opens you select Lucene Search. Note that Lucene is a free Java based information retrieval search tool that is particularly good at locating similar documents.

You can enter a standard Boolean search, and then click Apply to put it in the conditions list.

One of the most unique features of Lucene search is that it lets you specify the fuzziness level of your search in the search syntax. Numbers between 0 to 2 can be entered after a ~ to specify how many character terms in the results should differ from the search term. kCura's guide notes that 80% of misspellings have an edit distance of 1. So a search for:

treasure~1

. . . will find treasore, but not tresore.

A variation of this is a proximity search in which the syntax specifies the distance between two terms. So a search for

"treasure box"~2

. . . will find "treasure in a box" and "box of treasure", but not "treasure of priceless diamonds in a box"