turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

- JMP User Community
- :
- Discussions
- :
- Discussions
- :
- Importing Data

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Feb 21, 2011 2:57 AM
(1050 views)

Hello,

I'm using JMP 8 and I have a question: is it possible to import an already existing similarity matrix so that I can I apply clustering to it? As far as I understand, JMP handles the data in the table as individual datapoints and then applys a distance function of them to calculate for example the distance between data point A and B. But I have already the distances between these points and it is not possible to express them as individual values of A and B. So, can I import such an matrx and then apply just the clustering algorithms on it?

I'm using JMP 8 and I have a question: is it possible to import an already existing similarity matrix so that I can I apply clustering to it? As far as I understand, JMP handles the data in the table as individual datapoints and then applys a distance function of them to calculate for example the distance between data point A and B. But I have already the distances between these points and it is not possible to express them as individual values of A and B. So, can I import such an matrx and then apply just the clustering algorithms on it?

3 REPLIES

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Feb 21, 2011 10:44 AM
(923 views)

Hi Thomas,

There's no option in JMP 8 to use a matrix exactly as you describe. However, our experience is that using a similarity or distance matrix as input to the clustering platform as if it was raw data will generally give similar results results to what you would have gotten with the raw data. We do that frequently in our JMP Genomics product.

JMP 9 will take a distance matrix directly in the Clustering platform.

I hope that helps.

Jeff

There's no option in JMP 8 to use a matrix exactly as you describe. However, our experience is that using a similarity or distance matrix as input to the clustering platform as if it was raw data will generally give similar results results to what you would have gotten with the raw data. We do that frequently in our JMP Genomics product.

JMP 9 will take a distance matrix directly in the Clustering platform.

I hope that helps.

Jeff

-Jeff

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Feb 22, 2011 1:29 AM
(923 views)

Thanks for your answer. Perhaps, can you point me to an example or other ongoing documentation? I'm not sure how the similarity matrix should look like when I open it for clustering?

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Sep 26, 2012 2:14 PM
(923 views)

Hi Jeff,

I have a similar situation and I have JMP9.

Basically I have a distance matrix of N*N and I opened it in JMP. Then I tried to use the Analyze->Multivariant -> Cluster. Then I clicked the Data is Distance Matrix and clicked OK button. Then nothing happened. Exactly how do I do it?

Thanks,

Heng