• Post Reply Bookmark Topic Watch Topic
  • New Topic

why the ready tools are so fast?  RSS feed

 
sahar eb
Ranch Hand
Posts: 38
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
hi,
can somebody tell me why the ready tools are so fast? how they make their code so fast?
ok, let me explain more: I am trying to do some natural language processing on webpage texts. I am using OPEN NLP tool as base, and I add some small functions to it in order to improve. There is also some tools that do that, like Reverb, which is also built on the top of OPEN NLP. now, the interesting thing is that when I run my program; or only the basic part which is pure OPEN NLP, it takes longer than Reverb!!! Although Reverb is built on the top of OPEN NLP. So I am guessing they are doing something in Reverb to make every thing faster. I want to know what are those tricks? how can I make my code as fast as them?

Thanks!
 
Jesper de Jong
Java Cowboy
Sheriff
Posts: 16060
88
Android IntelliJ IDE Java Scala Spring
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
What are the "ready tools"?
 
Carles Gasques
Ranch Hand
Posts: 199
1
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Are you sure that Reverb uses parsing routines from OpenNLP and not from Guava?
 
sahar eb
Ranch Hand
Posts: 38
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
well they say Reverb is built on the top of OPEN NLP; this snippet is copied from Reverb website:
"ReVerb uses the following code and data, which are included in the release:
OpenNLP, including trained models
Weka
Guava
Apache Commons
A subset of pages from Wikipedia"
So there, it even uses the models from OPENNLP(which take alot of time to load), but it is still faster than OPENNLP. back to my question, I know one way to increase the execution time is to convert the data into binary, and then run the process on binary data. But any other Idea other than this?

beside that I need it, I am really curious to know more about this.

Thanks!
 
  • Post Reply Bookmark Topic Watch Topic
  • New Topic
Boost this thread!