5

I'm looking for tools I can use to perform text summarization. Specifically something I could use to summarize personal communication (i.e. email).

Preferably I'm looking for something open source and cross platform (Java?), but I'm open to any suggestion regardless of language or cost.

Any primers/hints on specific techniques or approaches that might work best would also be appreciated.

  • 1
    As I recall, Microsoft Word includes a text summarization tool that does a surprisingly reasonable (though by no means perfect) job, with adjustable parameters. – john lawler in exile Mar 9 '15 at 16:01
  • Thanks @johnlawlerinexile, but I'm looking for a library I can use in making my own application. – Fozefy Mar 9 '15 at 17:45
  • I concur with curiousdannii. This question does not appear to concern linguistics. – James Grossmann Mar 9 '15 at 23:35
  • That's fair, I did try there first, but didn't get any responses. – Fozefy Mar 10 '15 at 13:35
  • 2
    I'm voting to close this question as off-topic because it belongs on Software Recommendations. – curiousdannii Feb 7 '17 at 7:51
1

Here's a tool I happened upon recently: Context-controllable Content Summarization

And this looks to be related background information:

| improve this answer | |
  • Thanks for the attempt @arjan, but this isn't quite what I'm looking for. The sources at genopharmix.com are a bit too specific to 'life sciences', the tool provided allows you to specify context, but that can't be a requirement when the targeted area is personal communication. It could contain info for any context. The second source, Word2Vec, is interesting, but I'm hoping to find something more 'out of the box'. For example libots.sourceforge.net or github.com/Rotten194/summarize.py. Those are just a bit out-dated/primitive, so I'm hoping to find something more recent. – Fozefy Mar 9 '15 at 17:55
1

You do not have to do much if you are going to make an Extractive summarization.

  1. Do stop-word Removal

  2. Do stemming (optional)

  3. Count all words occurrences

  4. Extract sentence(s) that contains the word with most frequency in the text

This is the basic of text summarization which suits for emails very well, I guess.

| improve this answer | |
0

You could get your summaries by using an API. It's easy to integrate in almost any app. I use this and it works.

https://market.mashape.com/textanalysis/text-summarization

You send the text and the number of sentences you want returned. You can even try it to see if it works for you. Here's the address where you can evaluate it.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.