lebron letchev
asked on
Deep Document Summarization: Combining Text and Images
Dear all,
I would like to hear from you some ideas on how make "Deep Document Summarization: Combining Text and Images" besides extracting paragraphs. And if PHP it is an appropriate language for doing this task
I am looking for ideas and suggestions
Sincerely
I would like to hear from you some ideas on how make "Deep Document Summarization: Combining Text and Images" besides extracting paragraphs. And if PHP it is an appropriate language for doing this task
I am looking for ideas and suggestions
Sincerely
Julian Hansen
Can you explain a bit more what you mean by "Deep Document Summarization"?
ASKER
Its a summarization that combines texts and images.
Most text summarization methods are related to contextual information to identify relevant sentences for a summary. However, a many document contains , in addition to text, graphical elements, in general relevant, but many times, not necessary for our understanding of the contents the document.
Thank you
Most text summarization methods are related to contextual information to identify relevant sentences for a summary. However, a many document contains , in addition to text, graphical elements, in general relevant, but many times, not necessary for our understanding of the contents the document.
Thank you
That does not really make it clearer about what you are trying to do. Do you have a pattern for this and you need to implement in PHP or are you asking for a pattern on how to do this?
Your question is far to abstract to be able to give an answer - not sure what direction we are going.
If you can narrow it down in terms of what it is you specifically want to know that would help.
Your question is far to abstract to be able to give an answer - not sure what direction we are going.
If you can narrow it down in terms of what it is you specifically want to know that would help.
ASKER
I will be happy to know how to do this:
Not sure how to respond - I was asking for clarification - please read my post again.
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
Thank you so much.
This was the answer I wanted to hear. The image detection would not only be by image title, but deeper, as well you write "look at the pixels and determine that the image contains a dog, and then using NLP to look for content within the document that relates to dogs. "At this level, JAVA will be the best. And "it's not my cup of tea"
This was the answer I wanted to hear. The image detection would not only be by image title, but deeper, as well you write "look at the pixels and determine that the image contains a dog, and then using NLP to look for content within the document that relates to dogs. "At this level, JAVA will be the best. And "it's not my cup of tea"