link-grammar
link-grammar copied to clipboard
opencog
The CMU Link Grammar natural language parser
The idea that I tried appears in section 5 of: Li, H. (2016) Binary Tree’s Recursion Traversal Algorithm and Its Improvement. Journal of Computer and Communications, 4, 42-47. http://dx.doi.org/10.4236/jcc.2016.47006 I...
I started this discussion in PR #818, and I transfer it to here, with expansions. The purpose of the idea discussed here is to increase the speed and quality of...
In issue #743 I said: >extending idiom definitions to allow `_` and did not tell for what it may be good. Here is the story. From README.md: > ### "holes"...
Until and including 5.3.13, the library used the empty-word device. We already know that the linkages null-count (when it is >0), for sentences that got tokenized using empty-words, may often...
This is a meta-issue, design-change request, to treat capitalization (and possibly other things) as a kind-of pseudo-morphology. See issue #42 for context. The general issue is about refomulating tokenization (and...
I tried to add `ID*` to the length_limit of 1 (after of course allowing this usage of ID). I did this is order to check whether it can speed up...
Is there a reason the linkage count is 64 bits, considering it gets truncated to `INT_MAX` in `count.c`, and values greater than 24 bits are considered as `PARSE_NUM_OVERFLOW`? I ask...
An ongoing discussion of tokenization and its relation to morphlogy is happening in issue #42 It really should be happening here.
The following test works (but it needs a definition modification in order to actually be used): 4.0.regex: ``` NUMBER-GRP-g0: /^[1-9][0-9]?$/ NUMBER-GRP: /^[0-9]{3}$/ ``` 4.0.dict: ``` 1.g0 NUMBER-GRP-g0: ZZNS+ or NUMBERS;...
Currently, subscripted idioms are forbidden, so a dict entry like `a_b.c: something;` is considered to be a definition for the word `a_b` (a word which includes an underbar). This can...