2

I'm through a project which is about text simplification, there are several open sources which provide the parser of text such as Stanford parser. wondering if there any parser which is able to parse a text using machine learning!

mbatchkarov
  • 15,487
  • 9
  • 60
  • 79
S Gaber
  • 1,536
  • 7
  • 24
  • 43
  • 1
    What kind of simplification are you trying to do? What measures of complexity are you using? – dmh Jul 09 '12 at 16:37
  • I'm trying to do syntactical simplification using 'ML'. the measure of complexity is Analyzing complexity. I got documents which contain user requirements, i need to change the text into sentences, so those sentences can represent Story Cards as it's in "Extreme Programming" Agel methodology – S Gaber Jul 10 '12 at 01:56
  • Could you give us a concrete example of your input and desired output? "Analyzing complexity" doesn't seem well-defined to me, either. Determining the difficulty of a text is a non-trivial task; what sort of user-requirements do you have? – dmh Jul 10 '12 at 19:25
  • Maybe my answer [here](http://stackoverflow.com/a/9606606/547040) can help your problem – Khairul Jul 25 '12 at 08:40
  • @Khairul but that example doesn't use ML ... do u think it does !!! – S Gaber Aug 13 '12 at 02:22

1 Answers1

1

OpenNLP is a very good open source library for this. You can train this library to generate a model specific to your needs using many specific machine learning algorithms such MaxEnt. You can also see this discussion.

While OpenNLP is with Apache license, there is another tool called LingPipe which also provides similar functionality but different license.

Mahout is also a good library for machine learning, however, for that you need extract features from text using some other parser or your own implementation. Mahout needs features for learning.

Tutorials: 1. OpenNLP Tutorials 2. LingPipe Tutorials

vikas
  • 1,535
  • 1
  • 13
  • 22
  • tanks vikas, would you share any paper or tutorial that shows how those open source do it plz!! – S Gaber Jul 08 '12 at 08:45
  • These are very popular projects so finding tutorials should not be a big task. Still I have added some links to my answer. Also, added one more machine learning library. Hope this helps. – vikas Jul 08 '12 at 11:56
  • OpenNLP moved from SourceForge to Apache. Please refer to a newer documentation: http://opennlp.apache.org/documentation.html – wcolen Jul 10 '12 at 12:01
  • Maybe this OpenNLP works with Spanish language? If not....theres an optional component focused on spanish language??? Thanks – xavendano Jun 06 '13 at 16:15