FortisMain Page | About | Help | FAQ | Special pages | Log in

Internal Repetitions

From Fortis

A detailed explanation of how internal repetitions are calculated and measured. There is also a short description given on how to calculate any leverage that is gained through repetition.

Contents

How it works

The redundant segment analysis operates on a segment by segment basis to identify redundant segments during import. The various tables that are presented are all independent statistics generated from the file import. It is possible, but not straight forward, to add and subtract certain cells from the tables to get some additional useful information that is not presented. Only segments that are not pre-translated are considered in the tables. Some versions of Fortis generate a table for each file in a batch import as well as an aggregate for the entire import. When using the numbers you should base them on either the detailed report or the aggregate report, but not both. (The results will be different due to redundancies that occur between files but not within files.) Because this is confusing we may have removed the per-file statistics reporting from recent versions.

Two types: Complete and Partial redundancy

There are two different ways that a segment can be considered redundant. One is a complete redundancy, this is when the two segments are identical, with a possible exception for small variations such as the spacing before or after the segment. The other type of redundancy is a partial redundancy. This occurs when a segment text is the same, but the type or placement of tags in the segment differs.

Two counts: Distinct and Cumulative

Each report also includes a distinct count and a cumulative count. The numbers in the distinct column are a count of the number of occurences of each repeated segment, not counting the duplicates. The numbers in the cumulative column count the number of occurences including duplicates.

The counts are presented that tally totals for number of segments, number of words, and number of characters.

Finding Possible Leverage

If you take the cumulative repeated words and subtract the distinct repeated words this represents the amount of leverage you can get due to redundancy. The equation gets more complicated when you want to consider leverage due to full redundancy and then additional leverage due to partial redundancy.

One way to compute this is to find the following: LEVARGE FROM FULL REDUNDANCY = (CUMULATIVE FULL - DISTINCT FULL) LEVERAGE FROM PARTIAL AND FULL REDUNDANCY = (CUMULATIVE PART - DISTINCT PART) ADDITIONAL LEVERAGE DUE TO PARITAL REDUNDANCY = LEVERAGE FULL - LEVERAGE PARTIAL AND FULL

You could then multiply these numbers by the appropriate billing rates.

It is also possible to divide the word and segment counts to find measurements such as the average length of a redundant segment, etc.

Retrieved from "http://starfish.multiling.com/wiki/index.php/Internal_Repetitions"

This page has been accessed 319 times. This page was last modified 12:46, 24 February 2010.


Find

Browse
Main Page
Recent changes
Random page
Help
Edit
View source
Editing help
This page
Discuss this page
Post a comment
Printable version
Context
Page history
What links here
Related changes
My pages
Log in / create account
Special pages
New pages
File list
Statistics
Bug reports
More...