extract topic

m_keshavarz_com

Hello
Thank you
First of all, I searched but did not get the result
I want a review about the movie
Extract title
I clustered the data first and I want to find the title of each cluster, but I do not know the lda operator how to cluster?
And
I used a large datastore to reduce the size of svd. The operation takes an hour
How do I save the result and use the next loads for clustering?


Thank you
  With respect

sgenzer

hello @m_keshavarz_com - welcome to the community? Can you please post your process as per the guidelines so we can see what you are doing?

Scott

 

m_keshavarz_com

Hello
I did that, but it has an error
I want the sentences of a cluster by the filter operator
I separated.
What are their content?
What is clustering about?
what do I do?
Thanks if you help
With respect

sgenzer

hello @m_keshavarz_com so yes I would assume your process has an error or you would not be asking a question. This is why I'd like to see your process.

 

Scott

 

m_keshavarz_com

Hello
Sorry
I do not have access to me
I also changed the screen resolution but the xml menu was not visible
Then i can not send
look
I have a simple datasheet containing sentences and I want to cluster them according to the same titles?
But I do not know how and how can I do the operators?

Thanks
With respect

MartinLiebig

Hi,

The Input of LDA is a Tokenized collection of documents NOT the result of Process Documents. PLEASE have a look at the tutorial process which is added to the LDA operator. 

Best,

Martin

m_keshavarz_com

Hello
Yes I did
Thankful
Just how to find the optimal number of topics?
And what about?
And how do I understand which new entry text is placed on?

 

How to cluster the threads based on the same content

؟

thanks a lot