I am referring to the site deeplearning.net on how to implement the deep learning architectures. I have read quite a few research papers on document summarization (both single document and multidocument) but I am unable to figure to how exactly the summary is generated for each document.
Once the training is done, the network stabilizes during testing phase. So even if I know the set of features (which I have figured out) that are learnt during the training phase, it would be difficult to find out the importance of each feature (because the weight vector of the network is stabilized) during the testing phase where I will be trying to generate summary for each document.
I tried to figure this out for a long time but it's in vain.
If anybody has worked on it or have any idea regarding the same, please give me some pointers. I really appreciate your help. Thank you.
Best Answer
By themselves deep networks are not summarization algorithms, but they can help find key sentences that between them do summarize. The network is a model of text that you are likely to see in the corpus. The sentences you want in a summary should be typical of the corpus but different from each other so that between them a small set of sentences "covers" most of the corpus. So you want sentences that have a high probability according to the network individually, but also have low pointwise mutual information. PMI can be estimated by the network as well. I do not know how well this approach would compare with state of the art summarization.