[developers] DELPH-IN "stack"

[Stephan Oepen]

i added a pointer to the available information, though i also
want our entry page for (candidate) first-time users to only
contain as much information as is really needed to arouse
interest (and to enable a successful end-to-end run).

cheers, oe


On Wed, Jan 18, 2012 at 17:14, Emily M. Bender <ebender at uw.edu> wrote:
> Good to know.  I think it might be useful to have a note to that effect on
> that page (shall I WP:BeBold or leave it to you?), plus a link to info (if
> it's available) on how to hook up external NERs, should the user have
> one they would like to use.
>
> And agreed that alternative delivery modes should be documented :)
>
> Emily
>
> On Tue, Jan 17, 2012 at 10:45 PM, Stephan Oepen <oe at ifi.uio.no> wrote:
>> the :erg+tnt cpu definition activates all we have as of yet: unknown word handling (based on TnT PoS tags) and light-weight NE recognition via RE-based token mapping.
>>
>> come to think of it, i should maybe add a section on alternative delivery modes: HoG and ACE come to mind; any others that are in active (if maybe not quite broad) use currently and freely available?
>>
>> cheers, oe
>>
>>
>>
>> On 18. jan. 2012, at 02:55, "Emily M. Bender" <ebender at uw.edu> wrote:
>>
>>> Thanks, Stephan. That looks very helpful.  I will send the link
>>> to the folks at Boeing.  I suspect that in their intended use, they
>>> will need to contend with named entities, unknown word handling,
>>> etc.  How much of that is rolled into the latest and greatest
>>> LOGON?
>>>
>>> Emily
>>>
>>> On Tue, Jan 17, 2012 at 11:30 AM, Stephan Oepen <oe at ifi.uio.no> wrote:
>>>> hi emily,
>>>>
>>>> good to hear there is interest in our technology among the
>>>> second-largest producers of civil aircraft.  my apologies for
>>>> not getting back to you earlier, but i just composed a short
>>>> set of instructions for first-time use of the ERG.  see:
>>>>
>>>>  http://moin.delph-in.net/ErgProcessing
>>>>
>>>> i tested this (using my relatively vanilla account at a major
>>>> US university in the northwest), though it would of course
>>>> be great if this core functionality were validated by others.
>>>> i will be grateful for suggestions for improvement!
>>>>
>>>> looking at it now, i wonder why some folks still think it is
>>>> difficult to get going with the DELPH-IN toolchain.
>>>>
>>>> best wishes, oe
>>>>
>>>>
>>>> On Thu, Jan 12, 2012 at 05:20, Emily M. Bender <ebender at uw.edu> wrote:
>>>>> Dear all,
>>>>>
>>>>> At LSA Stephen Poteet from  Boeing was asking me about parsers,
>>>>> in the context of some IE work they are doing.  I suggested that he
>>>>> look into DELPH-IN tools, but I realized I don't know where to point
>>>>> him (or anyone) for the best way to set up a DELPH-IN based
>>>>> parsing stack (in this case, for English).  Do we have such documentation
>>>>> somewhere?  If not, something that could serve as a starting point?
>>>>>
>>>>> Thanks,
>>>>> Emiliy
>>>>>
>>>>> --
>>>>> Emily M. Bender
>>>>> Associate Professor
>>>>> Department of Linguistics
>>>>> Check out CLMS on facebook! http://www.facebook.com/uwclma
>>>
>>>
>>>
>>> --
>>> Emily M. Bender
>>> Associate Professor
>>> Department of Linguistics
>>> Check out CLMS on facebook! http://www.facebook.com/uwclma
>
>
>
> --
> Emily M. Bender
> Associate Professor
> Department of Linguistics
> Check out CLMS on facebook! http://www.facebook.com/uwclma



-- 
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+++ Universitetet i Oslo (IFI); Boks 1080 Blindern; 0316 Oslo; (+47) 2284 0125
+++    --- oe at ifi.uio.no; stephan at oepen.net; http://www.emmtee.net/oe/ ---
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++