Hi I have some data in excel that I want to convert to CSV in order to create a network based on mortality records
Here is what my data looks like
A) ID: 1,2,3 ... n
B) Sex: 1,2
C) Age: 73, 69, 45 ... n
D) Ethnic group: 1,2,3
E) Cause of Death (ICD-10 code): C45, E10 - I21 - J13, I21 - D53, ... n
A-E are my columns and the data you see are entries in raws
Column A has an ascending value (ID)
Column B is a binary value for Sex (Male, Female or 1,2)
Column C are age values
Column D has 3 ethnic groups (1,2,3)
and
Column D is the cause of Death: this is where I want to base the network. In this case there may be more than one entries in every raw...
So the idea is to link the data is the following so that e.g. if case 4 and case 17 share a common cause of death they will link-up (nodes by an edge)
Do you thinnk is feasible?
How do I arrange my data to get to this kind of network?
Thank you...
Network of co-morbidities
-
- Posts:68
- Joined:21 Jan 2012 18:34
- Location:Washington, DC, USA [phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1275: count(): Parameter must be an array or an object that implements Countable
Re: Network of co-morbidities
Hi Nikolaid,
That sounds like an interesting project.
One of the first things you'd want to decide is which values you want to be nodes and which values you want to be "properties" or "attributes" of a node. Some of your data could be used as either nodes or node attributes. You might choose to build only one network model that presents the data in a specific way; or, you might choose to build several models that present the data in different ways, thus allowing you to perform different types of analysis.
I think that you'd want to build a data set in which each individual was represented by a node with gender, ethnic group, and age as attributes for those nodes. Then, you'd probably want to have a node for each ICD in your data set. Then, you could create an edge linking each person node to the relevant ICD node.
Let me know if this helps. If you need help, I might be able to assist. You can contact me via this forum or use this platform to send me an email.
Good luck!
Francis
That sounds like an interesting project.
One of the first things you'd want to decide is which values you want to be nodes and which values you want to be "properties" or "attributes" of a node. Some of your data could be used as either nodes or node attributes. You might choose to build only one network model that presents the data in a specific way; or, you might choose to build several models that present the data in different ways, thus allowing you to perform different types of analysis.
I think that you'd want to build a data set in which each individual was represented by a node with gender, ethnic group, and age as attributes for those nodes. Then, you'd probably want to have a node for each ICD in your data set. Then, you could create an edge linking each person node to the relevant ICD node.
Let me know if this helps. If you need help, I might be able to assist. You can contact me via this forum or use this platform to send me an email.
Good luck!
Francis
Re: Network of co-morbidities
Hi,
Following up on Francis' suggestions, you might also find this software useful:
https://github.com/seinecle/Gaze/wiki/G ... r-networks
Best,
Clement
Following up on Francis' suggestions, you might also find this software useful:
https://github.com/seinecle/Gaze/wiki/G ... r-networks
Best,
Clement