5.9.1.2. Mode 2: We have a reference Table

<< Click to Display Table of Contents >>

Navigation:  5. Detailed description of the Actions > 5.9. Text Mining > 5.9.1. Correct Spelling (High-Speed action) >

5.9.1.2. Mode 2: We have a reference Table

 

Let’s go back the “city name” example: i.e. We now have two tables that contains city names:
 

The first table contains no errors inside the city names: This is the reference table.
 

The second table contains city names that might have some spelling mistakes and we want to correct these mistakes.

 

 
We’ll typically have such a graph:

 

clip0239

 

As input to the ANATEL~3_img694 CorrectSpelling Action, we have such a table:

 

clip0240

 

All the cities from the reference table have a “heavy weight” (greater than the parameter P3) such that no attempt to “correct” them will be made.

 

 

When using a reference table, the parameters from the ANATEL~3_img694 CorrectSpelling Action are, typically:

 

clip0241

 

 

When using a reference table, we’ll always have:
 

Parameter P4 is “Only regroup with “Heavy” Modalities

Parameter P5 is checked

Parameter P6 is equal to 1