Comments on: The Noob’s Guide to Parsing http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/ Wanderings into computational linguistics, science, social media and life... Mon, 23 Apr 2012 02:02:52 +0000 hourly 1 http://wordpress.com/ By: Ramkumar http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/#comment-3278 Mon, 20 Dec 2010 09:17:23 +0000 http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/#comment-3278 hey! know any methods to extract triples from the stanfordNLP ‘s output tree?

]]> By: maryN00b http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/#comment-1722 Wed, 03 Jun 2009 10:11:31 +0000 http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/#comment-1722 Hello! can someone please help me with a assignment I`m worikng on! I`m supposed to Construct a formal grammar for capturing the structure of names and then write a parser for English names by encoding your grammar in prolog using the DCG mechanism.

I`m very much a noob in this!

]]> By: DegreeStudent http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/#comment-1597 Sun, 22 Mar 2009 07:28:25 +0000 http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/#comment-1597 Hi,

First i would like to thank you for this helpful post. I need a help regarding stanford-parser.

i am doing a project and one of the the things i need to do is parsing (taggin each word in a text file). I have downloaded the stanford-parser zip file and could run the gui file. However, my project is developed in NetBeans IDE 6.1. I would appreciate if you tell me how to add the parser to my project and how to import the classes. I also need the output to be in the form where each word appear withe its tag next to it, and not in the tree form…

I hope i can get what i am looking for here …

thank you,

]]> By: Jason Adams http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/#comment-1128 Fri, 05 Dec 2008 04:53:53 +0000 http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/#comment-1128 Awesome! Thanks for letting me know.

]]> By: Jeremy http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/#comment-1127 Fri, 05 Dec 2008 04:47:02 +0000 http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/#comment-1127 Jason, thank you for your writeup. It’s helped me get started with a school project I’m working on, which is providing a “deep analysis” of a parser. Unfortunately I have almost no natural language processing knowledge, so this primer has been a big help to getting me started in at least understanding the basics of what I’ll have to analyze.

]]> By: Jason Adams http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/#comment-1088 Fri, 07 Nov 2008 03:32:15 +0000 http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/#comment-1088 I’m not sure exactly what you mean by the “limitations created by parsers.” Do you mean their inability to recognize certain things?

Statistical parsers are inherently flawed, though. In order to be computationally tractable you have to make a number of assumptions that are just plain wrong. The trick might be to figure out a way of incorporating world knowledge without blowing up the computation and taking forever to build. Or maybe it will take a completely different approach to solve… Who knows? :)

]]> By: shaun http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/#comment-1087 Fri, 07 Nov 2008 00:51:50 +0000 http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/#comment-1087 I am surprised at the limitations created by parsers generally. They seem to be merely scratching the surface of language, almost as if we are approaching things from the wrong perspective. We probably need some Einstein to come up with an entirely new way of looking at the issue of parsing.

]]> By: Jason Adams http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/#comment-974 Sat, 02 Aug 2008 13:39:04 +0000 http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/#comment-974 I don’t have first-hand experience with anything, since the only prolog I’ve written was a homework assignment in an AI class. A quick hunt on the Googles finds a couple things though, that may be of interest:

Pronto: http://www.ai.uga.edu/mc/ProNTo/

Book on NLP with Perl and Prolog: http://www.cs.lth.se/home/Pierre_Nugues/ilppp/slides.shtml

Anyhow, good luck. Might want to try asking on the comp.ai.nat-lang newsgroup: http://groups.google.com/group/comp.ai.nat-lang/topics?hl=en

]]> By: Hok http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/#comment-973 Sat, 02 Aug 2008 13:28:44 +0000 http://mendicantbug.com/2007/10/25/the-noobs-guide-to-parsing/#comment-973 Hi- I’m glad that I’ve found this post, because it seems very helpful for my experimental chatterbot project- it’s called ANNA, written in Prolog.
The plan is, I need it to be adaptive.
It currently would take Natural Language as a ‘sense’- i.e. the only stimuli she’ll ever receive are strings from the keyboard. Therefore, she would have to start off with quite a lot of innate knowledge about the English language and how to interpret it correctly and assimilate, or reject incoming propositions, keeping, all the while, a consistent model of the world within her knowledge base.
At the moment, I am in need of a way of tagging parts of speech, so that she could identify just what a sentence ‘means’ rather than simply extracting word after word and rearranging them.
But, for her to be fully ‘adaptive’, it is important that the parser, as well as the ‘semantics extraction’ part (after POST), also expressed in Prolog. She must be able to reason about the way she does things, and she can, only by using Prolog’s meta-programming facilities. So, I wonder if you know of any good Prolog-based POST libraries that require little or no external files, and yet provide a good English PCFG parser?
I fear I may well have to roll my own, at the expense of a lot of time and re-inventing the wheel.
Cheers!

]]>