Welcome,
Guest
. Please
login
or
register
.
Did you miss your
activation email?
Home
Help
Search
Login
Register
Rapid-I
Rapid-I Forum
»
RapidMiner
»
Data Mining / ETL / BI Processes
»
is the order of attributes when doing a similarity comparison?
Pages: [
1
]
« previous
next »
Print
Author
Topic: is the order of attributes when doing a similarity comparison? (Read 172 times)
wasperen
Newbie
Posts: 16
is the order of attributes when doing a similarity comparison?
«
on:
February 07, 2012, 12:02:01 AM »
Hi all,
When I have two example sets that contain the same attributes but in a different order... Do similarity computations take that into account or are they relying on the fact that attributes are in the same order?
Or, does [aap=1, noot=0] compared to [noot=0, aap=1] give the Euclidian distance zero or srt(2)?
thanks,
Willem
Logged
awchisholm
Full Member
Posts: 157
Re: is the order of attributes when doing a similarity comparison?
«
Reply #1 on:
February 08, 2012, 06:57:10 AM »
Hello Willem
The last time I tried this I found that the order does matter.
http://rapidminernotes.blogspot.com/2011/05/care-needed-using-cross-distances.html
regards
Andrew
Logged
Pages: [
1
]
Print
« previous
next »
Jump to:
Please select a destination:
-----------------------------
General Community
-----------------------------
=> News and Updates
=> Data Mining
=> Chit Chat
-----------------------------
RapidMiner
-----------------------------
=> Getting Started
=> Data Mining / ETL / BI Processes
=> Problems and Support
=> Feature Requests
=> Development
-----------------------------
RapidAnalytics
-----------------------------
=> Getting Started
=> Applications and Integration
-----------------------------
RapidNet
-----------------------------
=> Getting Started
=> Problems and Support
Loading...