Semantic Network Node Selection

[John Ravenel]

I am creating several semantic networks using AUTOMAP, which restricts the networks to 1999 nodes.  I just realized the program simply cuts the nodes off based on the alphabetic order (losing all concepts that start after the letter "d"). 

I have tried simply culling the texts by deleting all concepts below a certain frequency, but it did not eliminate the problem and still returned around 18,000 concepts, resulting in the same issue with the semantic networks being limited based on alphabetical ordering.   

Is there away to select the 1999 nodes using another metric other than alphabetic order, such as the top 1999 in terms of frequency?

[Kathleen Carley]

you need to create a list of all concepts and then give AutoMap a list of the concepts you want and tell it to code for only those

--
Computational Analysis of Social and Organizational Systems (CASOS)
The Institute for Software Research, Department of Computer Science
Carnegie Mellon University, Pittsburgh, PA, USA
http://casos.cs.cmu.edu/
---
You received this message because you are subscribed to the Google Groups "ORA Google Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ORA-google-gro...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ORA-google-group/d6cc04ea-dd26-42bc-8d08-1636203c0f69%40googlegroups.com.

[John Ravenel]

Thanks so much for getting back to me.  Where do I do that in terms of creating the semantic network?  

I can see how to create and cull concept lists (as well as develop delete lists and thesauri), just not how to tell Automap to code based only on certain concepts.  

On Tuesday, May 12, 2020 at 4:28:44 PM UTC-4, Kathleen Carley wrote:

you need to create a list of all concepts and then give AutoMap a list of the concepts you want and tell it to code for only those

On Tue, May 12, 2020 at 4:26 PM John Ravenel <j.bisho...@gmail.com> wrote:

I am creating several semantic networks using AUTOMAP, which restricts the networks to 1999 nodes.  I just realized the program simply cuts the nodes off based on the alphabetic order (losing all concepts that start after the letter "d"). 

I have tried simply culling the texts by deleting all concepts below a certain frequency, but it did not eliminate the problem and still returned around 18,000 concepts, resulting in the same issue with the semantic networks being limited based on alphabetical ordering.   

Is there away to select the 1999 nodes using another metric other than alphabetic order, such as the top 1999 in terms of frequency?

--
Computational Analysis of Social and Organizational Systems (CASOS)
The Institute for Software Research, Department of Computer Science
Carnegie Mellon University, Pittsburgh, PA, USA
http://casos.cs.cmu.edu/
---
You received this message because you are subscribed to the Google Groups "ORA Google Group" group.

To unsubscribe from this group and stop receiving emails from it, send an email to ORA-goog...@googlegroups.com.

[Kathleen Carley]

two approaches

Approach 1 is a bit of a hack, but it seems to achieve what you want:

1. Generate a concept list from the target text(s).
2. Open the concept list and cut out the concepts column and save as a single column delete list.
3. Edit the delete list, removing the concepts you want to keep in the text (I simply prepended an X to concepts).
4. Apply the delete list the the target text(s).
5. Generate the meta-network (using a dummy thesaurus, since one is required).

This gives you a selected concepts only meta-network.  

Approach 2, create a real thesaurus that includes only the concepts you want and all the variants being merged into them.  Then tell it to use only the concepts in the thesaurus

To unsubscribe from this group and stop receiving emails from it, send an email to ORA-google-gro...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ORA-google-group/3c9d4b3b-1ea8-45d9-a35c-2c279d9b5e13%40googlegroups.com.

[John Ravenel]

That is great.  Thank you for your help.  

On Tuesday, May 12, 2020 at 7:26:10 PM UTC-4, Kathleen Carley wrote:

two approaches

Approach 1 is a bit of a hack, but it seems to achieve what you want:

1. Generate a concept list from the target text(s).
2. Open the concept list and cut out the concepts column and save as a single column delete list.
3. Edit the delete list, removing the concepts you want to keep in the text (I simply prepended an X to concepts).
4. Apply the delete list the the target text(s).
5. Generate the meta-network (using a dummy thesaurus, since one is required).

This gives you a selected concepts only meta-network.  

Approach 2, create a real thesaurus that includes only the concepts you want and all the variants being merged into them.  Then tell it to use only the concepts in the thesaurus

To view this discussion on the web visit https://groups.google.com/d/msgid/ORA-google-group/3c9d4b3b-1ea8-45d9-a35c-2c279d9b5e13%40googlegroups.com.