Comments for OpenCog Brainwave http://brainwave.opencog.org The first ultraintelligent machine is the last invention that man need ever make. -I. J. Good, 1965 Thu, 08 Oct 2009 16:10:00 +0000 http://wordpress.com/ hourly 1 Comment on Hacking on Link-Grammar by jasonforceau http://brainwave.opencog.org/2008/08/17/hacking-on-link-grammar/#comment-15 jasonforceau Thu, 08 Oct 2009 16:10:00 +0000 http://opencog.wordpress.com/?p=41#comment-15 Hi, i am looking for tools for syntactic analysis for the system of my Final Year Project and found your post so interesting. But I don't know Link-Grammar. And how can I start up to using Link-Grammar to apply into my system? is it using c language? anyother language? Hi, i am looking for tools for syntactic analysis for the system of my Final Year Project and found your post so interesting. But I don’t know Link-Grammar. And how can I start up to using Link-Grammar to apply into my system? is it using c language? anyother language?

]]> Comment on Distribution of Mutual Information by linasv http://brainwave.opencog.org/2009/03/11/distribution-of-mutual-information/#comment-14 linasv Tue, 08 Sep 2009 18:38:26 +0000 http://brainwave.opencog.org/?p=89#comment-14 These graphs were discussed on the corpora mailing list in March 2009. See http://mailman.uib.no/public/corpora/2009-March/008193.html and followups in that thread. It was concluded that the gross features of the graph (i.e. the exponential fall-off) can be explained by drawing random pairs of words from a Zipfian distribution for single words (which seemed plausible to me, although I did not do the math to double-check/verify this). These graphs were discussed on the corpora mailing list in March 2009. See http://mailman.uib.no/public/corpora/2009-March/008193.html and followups in that thread.

It was concluded that the gross features of the graph (i.e. the exponential fall-off) can be explained by drawing random pairs of words from a Zipfian distribution for single words (which seemed plausible to me, although I did not do the math to double-check/verify this).

]]> Comment on Frequency of grammatical disjuncts by linasv http://brainwave.opencog.org/2009/07/06/frequency-of-grammatical-disjuncts/#comment-13 linasv Sat, 11 Jul 2009 00:58:18 +0000 http://opencog.wordpress.com/?p=123#comment-13 Martin Reynaert wrote to say: ''<em>From what I have learned from the work of mainly Ramon Ferrer i Cancho ( http://www.lsi.upc.edu/~rferrericancho/publications_by_year.html ), I would say that adding syntactic patterns to the words turns the natural language into more formal language. For more formal language a power law exponent well above 1 is `natural'.</em>'' Martin Reynaert wrote to say: ”From what I have learned from the work of mainly Ramon Ferrer i Cancho ( http://www.lsi.upc.edu/~rferrericancho/publications_by_year.html ), I would say that adding syntactic patterns to the words turns the natural language into more formal language. For more formal language a power law exponent well above 1 is `natural’.”

]]> Comment on Determining word senses from grammatical usage by OpenCog « Doppelleben http://brainwave.opencog.org/2009/01/12/determining-word-senses-from-grammatical-usage/#comment-12 OpenCog « Doppelleben Wed, 08 Jul 2009 10:21:09 +0000 http://brainwave.opencog.org/?p=70#comment-12 [...] I just discovered this cool blogpost on OpenCog. It is a very interesting blog by someone who has both time and talent for nice [...] [...] I just discovered this cool blogpost on OpenCog. It is a very interesting blog by someone who has both time and talent for nice [...]

]]> Comment on Frequency of grammatical disjuncts by ftyers http://brainwave.opencog.org/2009/07/06/frequency-of-grammatical-disjuncts/#comment-11 ftyers Wed, 08 Jul 2009 07:43:31 +0000 http://opencog.wordpress.com/?p=123#comment-11 Two comments really... the first is that perhaps the "resisting the use of formulaic sentences" comes from using more formal text as opposed to e.g. speech or chat type text. When I write for Wikipedia, I often find myself re-wording sentences. The second is, have you considered trying the same with the Persian Link grammar ? [1] 1. http://www.ling.ohio-state.edu/~jonsafari/#projects Two comments really… the first is that perhaps the “resisting the use of formulaic sentences” comes from using more formal text as opposed to e.g. speech or chat type text. When I write for Wikipedia, I often find myself re-wording sentences.

The second is, have you considered trying the same with the Persian Link grammar ? [1]

1. http://www.ling.ohio-state.edu/~jonsafari/#projects

]]> Comment on GSoC 2009 project list by The Singularity Institute Blog : Blog Archive : Google Summer of Code projects announced for OpenCog http://brainwave.opencog.org/2009/04/20/gsoc-2009-project-list/#comment-10 The Singularity Institute Blog : Blog Archive : Google Summer of Code projects announced for OpenCog Tue, 21 Apr 2009 10:34:50 +0000 http://brainwave.opencog.org/?p=110#comment-10 [...] From Dr. Joel Pitt, who is working on the OpenCog project with the sponsorship of the SIAI: [...] [...] From Dr. Joel Pitt, who is working on the OpenCog project with the sponsorship of the SIAI: [...]

]]> Comment on Fun with first-order inference by Visualization with UbiGraph « OpenCog Brainwave http://brainwave.opencog.org/2008/10/22/fun-with-first-order-inference/#comment-9 Visualization with UbiGraph « OpenCog Brainwave Thu, 12 Feb 2009 00:07:28 +0000 http://opencog.wordpress.com/?p=61#comment-9 [...] This follows on from the visualisation stuff I did for Ben’s presentation at the Singularity Summit demonstrating first order PLN inference on word pairs. [...] [...] This follows on from the visualisation stuff I did for Ben’s presentation at the Singularity Summit demonstrating first order PLN inference on word pairs. [...]

]]> Comment on Determining word senses from grammatical usage by Posts about Google as of January 12, 2009 | The Lessnau Lounge http://brainwave.opencog.org/2009/01/12/determining-word-senses-from-grammatical-usage/#comment-8 Posts about Google as of January 12, 2009 | The Lessnau Lounge Tue, 13 Jan 2009 01:07:42 +0000 http://brainwave.opencog.org/?p=70#comment-8 [...] could be helpful). Please send this information to sowehear AT gmail DOT com by Friday, 16 Determining word senses from grammatical usage - brainwave.opencog.org 01/12/2009 I’ve recently been tinkering with a mechanism for determining [...] [...] could be helpful). Please send this information to sowehear AT gmail DOT com by Friday, 16 Determining word senses from grammatical usage – brainwave.opencog.org 01/12/2009 I’ve recently been tinkering with a mechanism for determining [...]

]]> Comment on Hacking on Link-Grammar by linasv http://brainwave.opencog.org/2008/08/17/hacking-on-link-grammar/#comment-7 linasv Mon, 29 Sep 2008 19:58:12 +0000 http://opencog.wordpress.com/?p=41#comment-7 As of early summer, relEx has had the so-called "compact file format", and a bunch of text has been parsed: the entire simple-english wikipedia, a voice-of-america corpus, about 8 project gutenberg books (including war and peace) and decent chunk of the english wikipedia (parsing all of wikipedia will require about 10 cpu-years). These pre-parsed texts are available at http://relex.swlabs.org/~linas/ and a statistical package tailored to this is at https://launchpad.net/relex-statistical. As of early summer, relEx has had the so-called “compact file format”, and a bunch of text has been parsed: the entire simple-english wikipedia, a voice-of-america corpus, about 8 project gutenberg books (including war and peace) and decent chunk of the english wikipedia (parsing all of wikipedia will require about 10 cpu-years).

These pre-parsed texts are available at

http://relex.swlabs.org/~linas/

and a statistical package tailored to this is at

https://launchpad.net/relex-statistical.

]]> Comment on Hacking on Link-Grammar by jrowe47 http://brainwave.opencog.org/2008/08/17/hacking-on-link-grammar/#comment-6 jrowe47 Sat, 27 Sep 2008 20:46:54 +0000 http://opencog.wordpress.com/?p=41#comment-6 One idea that might speed things up would be to parse a crapload of arbitrary sentences (say, a book or ten) into grammatical units, so that you can see a general collection of valid sentences. You could also see general meta-relationships between sentence structures (the points where contextual data matter, and the points where grammar itself is providing a context.) Training a neural net to recognize and tag parts of sentences as related to their link-grammar categorization would also allow general assumptions to be made about content grammar relationships, and might give a jump up on hand-crafting more link-grammar entries. Eventually, a well trained neural net could easily tag arbitrary grammar/data/context relationships. One idea that might speed things up would be to parse a crapload of arbitrary sentences (say, a book or ten) into grammatical units, so that you can see a general collection of valid sentences. You could also see general meta-relationships between sentence structures (the points where contextual data matter, and the points where grammar itself is providing a context.)

Training a neural net to recognize and tag parts of sentences as related to their link-grammar categorization would also allow general assumptions to be made about content grammar relationships, and might give a jump up on hand-crafting more link-grammar entries. Eventually, a well trained neural net could easily tag arbitrary grammar/data/context relationships.

]]>