Mar
26
comment NLP: retrieve vocabulary from text
Yes, wiktionary is semi-structured, but you still can extract word forms from wiktionary definitions (here's some example code how you can process them - lisp-univ-etc.blogspot.com/2013/06/…, but you can also look into different tools like wiktionary-to-mysql, wiktionary-to-redis or wiktionary-to-dbpedia)
Mar
26
comment NLP: retrieve vocabulary from text
you can use wiktionary database for yiur first approach
Feb
26
awarded Enlightened
Feb
26
awarded Nice Answer
Feb
25
comment Using cl-glfw3 on windows
This is how CFFI searches for foreign linraries: common-lisp.net/project/cffi/manual/html_node/…
Feb
24
answered Reading from device files
Feb
18
comment NLP: How to correctly normalise a feature for gender classification?
@jablesauce yes, if you already have each frequency as a value between 0 and 100 and the sum of all frequencies is also below 100, you have the correct formula. So, just dividing by 100 will scale your feature into the 0-1 range
Feb
17
answered NLP: How to correctly normalise a feature for gender classification?
Feb
3
revised How does generative language model work in the natural language processing?
edited body
Jan
29
revised What's the simplest Common Lisp idiom for queued execution of function calls?
added 19 characters in body
Jan
29
answered What's the simplest Common Lisp idiom for queued execution of function calls?
Jan
12
answered Amending a condition as the stack unwinds in Common Lisp
Dec
26
answered Convert PCFG to CNF?
Dec
24
accepted Compile-file with in-memory data instead of real file in Common Lisp
Dec
24
comment nltk built-in (or easily obtainable) probabilistic parsing models?
@Yasen oh, I've found the dependencies in ch/ dir, thanks. Unfortunately, no depednencies for English are available (I guess, this also has to do with the PTB restrictions). However, on the Stanford website you can find the dependencies extracted from the English Web Treebank (www-nlp.stanford.edu/data/EWT_DependencyGoldStandard_v0.3.tgz)
Dec
23
comment nltk built-in (or easily obtainable) probabilistic parsing models?
@Yasen btw, I've checked the Univerasl dependency bank and haven't found the actual files with dependencies in it. Do you have a direct link?
Dec
23
comment nltk built-in (or easily obtainable) probabilistic parsing models?
@Yasen sorry, I didn't clarify that I meant the consituency treebanks (referred in the question as PCFGs). Also, Stanford parser is a commercial product and that's why they are able to pay for the treebanks license (not to mention the ties between Stanford and LDC). (The free version of the Stanford Parser doesn't permit commercial use)
Dec
23
comment nltk built-in (or easily obtainable) probabilistic parsing models?
In fact, all the corpora you could use for training English parsers are available only from LDC with a restrictive license. This is why it's very hard if not impossible for open and free parsers to provide models based on them.
Nov
16
comment how to develop a tool for phonological transformation purpose?
take a look at this: github.com/dmort27/HsSPE
Oct
3
awarded Yearling
1 2 3 4 5