Difference between revisions of "Matchcode Optimization"

From Melissa Data Wiki
Jump to navigation Jump to search
(Created page with "{{UnderConstruction}} {{MatchcodeOptimizationNav |MatchcodeOptimizationCollapse= }} Placeholder Category:MatchUp Hub Category:Matchcode Optimization")
 
Line 4: Line 4:
 
}}
 
}}
  
Placeholder
+
==Evaluate Matchcode==
 +
To help you understand the effect of constructing and implementing a sub-optimal matchcode, the Evaluate Matchcode feature evaluates the five critical areas of a matchcode which determine whether one can expect the best throughput when running a process. Evaluate matchcode will indicate if your matchcode is optimized, has a warning, or is sub optimal.
 +
 
 +
[[File:MCO_Evaluate_Optimized.png|link=]]
 +
 
 +
 
 +
[[File:MCO_Evaluate_Warning.png|link=]]
 +
 
 +
 
 +
[[File:MCO_Evaluate_Sub.png|link=]]
 +
 
 +
 
 +
==Five Critical Matchcode Areas==
 +
===First Component Used in All Set Columns===
 +
When the first component is not checked (not evaluated) in all used combinations, processing will be very slow, as matchkeys cannot be grouped by that first component into optimal neighborhood clusters.
 +
*[[Matchcode Optimization:First Component|First Component]]
 +
 
 +
===Fuzzy Settings===
 +
If a component uses a fuzzy algorithm other than exact, each record compared will produce speed penalty on keybuilding or on that algorithms required computation when deduping.
 +
*[[Matchcode Optimization:Fuzzy Algorithms|Fuzzy Algorithms]]
 +
 
 +
===Swap Matching===
 +
When each key is compared, swapping one records values (for example: an inverse name format) and reattempting to identify a match will produce slower processing speeds.
 +
*[[Matchcode Optimization:Swap Matching|Swap Matching]]
 +
 
 +
===Blank Matching===
 +
Allowing one of the 2 values being compared to either be blank or a single initial will require the matchup engine to make an extra comparison(s) and will produce a speed penalty.
 +
*[[Matchcode Optimization:Blank Matching|Blank Matching]]
 +
 
 +
===Advanced Matchcode Component Type===
 +
Most matchcode component types simply specify how to extract source data into a records matchkey, but Advanced types will also tell the deduping engine to perform a calculation foreach record compared, even if the Distance property is set to zero.
 +
*[[Matchcode Optimization:Advanced Component Types|Advanced Component Types]]
  
  
 
[[Category:MatchUp Hub]]
 
[[Category:MatchUp Hub]]
 
[[Category:Matchcode Optimization]]
 
[[Category:Matchcode Optimization]]

Revision as of 16:57, 20 September 2018

ICON WIKI Warning.png
This page is still under construction!
Melissa Data strives to give you the most complete and up-to-date information about our products as possible. To do this, we must maintain our documentation. This means the content may not be complete or correct. Use at your own risk!

← MatchUp Hub

Matchcode Optimization Navigation
Matchcode Optimization
First Component
Fuzzy Algorithms
Swap Matching
Blank Matching
Advanced Component Types
Algorithms
Accunear
Alphas
Consonants
Containment
Dice's Coefficient
Double Metaphone
Exact
Fast Near
Frequency
Frequency Near
Jaccard Similarity Coefficient
Jaro
Jaro-Winkler
Longest Common Substring (LCS)
MD Keyboard
Needleman-Wunsch
N-Gram
Numeric
Overlap Coefficient
Phonetex
Smith-Waterman-Gotoh
Soundex
UTF8 Near
Vowels


Evaluate Matchcode

To help you understand the effect of constructing and implementing a sub-optimal matchcode, the Evaluate Matchcode feature evaluates the five critical areas of a matchcode which determine whether one can expect the best throughput when running a process. Evaluate matchcode will indicate if your matchcode is optimized, has a warning, or is sub optimal.

MCO Evaluate Optimized.png


MCO Evaluate Warning.png


MCO Evaluate Sub.png


Five Critical Matchcode Areas

First Component Used in All Set Columns

When the first component is not checked (not evaluated) in all used combinations, processing will be very slow, as matchkeys cannot be grouped by that first component into optimal neighborhood clusters.

Fuzzy Settings

If a component uses a fuzzy algorithm other than exact, each record compared will produce speed penalty on keybuilding or on that algorithms required computation when deduping.

Swap Matching

When each key is compared, swapping one records values (for example: an inverse name format) and reattempting to identify a match will produce slower processing speeds.

Blank Matching

Allowing one of the 2 values being compared to either be blank or a single initial will require the matchup engine to make an extra comparison(s) and will produce a speed penalty.

Advanced Matchcode Component Type

Most matchcode component types simply specify how to extract source data into a records matchkey, but Advanced types will also tell the deduping engine to perform a calculation foreach record compared, even if the Distance property is set to zero.