Want to win a PS4? Go Premium and enter to win our High-Tech Treats giveaway. Enter to Win

x
Solved

# Euclidean distance between list of Vectors with different size of lists

Posted on 2011-03-01
Medium Priority
537 Views
Hello,
I have 2 lists
List1 = [ V1,V2,V3,...,Vn] and List2 = [V1,V2,V3,....,Vk]   ==> Where K != n
And Where all the vectors in both lists have same length of course.

How can i compute the distance between these 2 lists of vectors?
1. Shall i take the centrois of each lists?

Thank you.
0
[X]
###### Welcome to Experts Exchange

Add your voice to the tech community where 5M+ people just like you are talking about what matters.

• Help others & share knowledge
• Earn cash & points
• 9
• 3

Author Comment

ID: 35015076
say i take the centroids or center of gravity how can i compute it?
0

LVL 84

Expert Comment

ID: 35015082
That's one way.
Other ways include
Minimum distance from a vector in one list to a vector in the other list
Maximum distance from a vector in one list to a vector in the other list
Average distance from a vector in one list to a vector in the other list
Sum of distances from the centroid of both lists to the vectors in each list
...
What properties do you want this distance to have?
0

Author Comment

ID: 35015126
Actually i want to compute the dissimilarity between list1 and list2.
0

Author Comment

ID: 35015163
if i want to compute the center of gravity or centroid for a list of elements it's easy:
L = [e1,e2,...,en] it's easy just the mean.
But what is confusing me a little is this i can't figure it out:
If L = [v1,v2,....,vn] with each vi = [ x1,...xk] now how can i do it?? ouf it's a little complicated.
0

LVL 84

Expert Comment

ID: 35015189
for the centroid, you can take the centroid of each component.
What properties do you want the dissimilarity to have?
What kinds of lists do you want to be dissimilar to other lists?
For example, one property of the centroid measure you propose is that changing the order of the elements of a list does not change its dissimilarity.
Another is that it satisfies the triangle inequality.
0

Author Comment

ID: 35015229
Well actually it should satisfy the Triangle inequality.
0

Author Comment

ID: 35015233
well i have a question for the centroid but i'll ask it later.
0

LVL 84

Accepted Solution

ozo earned 1000 total points
ID: 35015256
Ok, but many kinds of dissimilarity measures satisfy the triangle inequality.
one advantage of centroid over some others is that it is fast and easy to compute
0

Author Comment

ID: 35015260
Sorry Correction, Correction
Well actually it should satisfy the Triangle inequality. NOT TRIANGLE INEQUALITY!!!
I want to have a Diagonal Matrix. all entries on the diagonal should be 0.
So i should compute the euclidean distance.

SORRY !!!!!!!!!!!!!
0

Author Comment

ID: 35015263
Ouf it's confusing now. Sorry for the mistake.
0

Author Comment

ID: 35015683
yeah well maybe i can work on the hausdorff distance i just find something on google. It's not bad and it's close to what u said. MAx[Min[d1,....dn]]

0

Author Closing Comment

ID: 35016666
no follow up sorry. I got the answer from a free forum.
Thank you anyway.
take care....
0

## Featured Post

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

One of Google's most recent algorithm changes affecting local searches is entitled "The Pigeon Update." This update has dramatically enhanced search inquires for the keyword "Yelp." Google searches with the word "Yelp" included will now yield Yelp aâ€¦
When there is a disconnect between the intentions of their creator and the recipient, when algorithms go awry, they can have disastrous consequences.
I've attached the XLSM Excel spreadsheet I used in the video and also text files containing the macros used below. https://filedb.experts-exchange.com/incoming/2017/03_w12/1151775/Permutations.txt https://filedb.experts-exchange.com/incoming/201â€¦
Sometimes it takes a new vantage point, apart from our everyday security practices, to truly see our Active Directory (AD) vulnerabilities. We get used to implementing the same techniques and checking the same areas for a breach. This pattern can reâ€¦
###### Suggested Courses
Course of the Month10 days, 21 hours left to enroll