Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[bioontology-support] Fwd: Proxy error when calling text mining service

Paul R Alexander palexander at stanford.edu
Mon Oct 18 14:34:38 PDT 2010


Marco,

Sorry for the delay in responding. I checked and it looks like there 
isn't a hard limit on the text size, just however long your timeout is 
set for. That being said, I was able to pass in chunks that were 8000 
characters and get a response back in a reasonable time.

Essentially this means that you can determine the size, you'll just need 
to determine how long you are willing to wait for a response and do some 
testing against the service to make sure it works.

Let me know if you still have questions.

Paul R Alexander
Web / UI Developer
NCBO BioPortal <http://bioportal.bioontology.org>
Stanford Center for Biomedical Informatics Research


On 10/14/10 1:51 PM, Marco Ocana wrote:
> Hi Paul,
>
> Thanks for looking into this.
> I have in the past gotten results for this document, but not sure what 
> were the circumstances.
> Do you guys have advice on on how big the chunks can be? If we need to 
> split up the document, then we would like to keep the chunks as large 
> as possible.
>
> Regards,
>
> Marco
> ------------------------------------------------------------------------
> *From:* Paul R Alexander <palexander at stanford.edu>
> *To:* Natasha F. Noy <noy at stanford.edu>
> *Cc:* support Support <support at bioontology.org>; Marco Ocana 
> <marco.ocana at balboasystems.com>; Paolo Ciccarese 
> <paolo.ciccarese at gmail.com>
> *Sent:* Wed, October 13, 2010 3:30:42 PM
> *Subject:* Re: [bioontology-support] Fwd: Proxy error when calling 
> text mining service
>
> Marco,
>
> I'm not an Annotator expert, but I did look into the problem and it 
> appears that the file is just too big to be processed in one call (it 
> timed out for me). I had no problem processing the entire thing so 
> long as I split it up into chunks that were generally 6000-8000 
> characters long.
>
> Is this test case something that you are normally able to process 
> without difficulty? There's a possibility that if you set your timeout 
> long enough it would eventually come back with a result. As far as I 
> know, the Annotator isn't built to handle text from an entire article, 
> which it appears this is (someone please correct me on this if I am 
> wrong).
>
> Let me know if this workaround is suitable for you.
>
> Best,
> Paul R Alexander
> Web / UI Developer
> NCBO BioPortal <http://bioportal.bioontology.org>
> Stanford Center for Biomedical Informatics Research
>
>
> On 10/13/10 10:29 AM, Natasha F. Noy wrote:
>> Marco,
>>
>> Forwarding this to support so that we can get more eyeballs on this. 
>> I myself don't know what is going on.
>>
>>
>> Natasha
>>
>>
>>
>> Begin forwarded message:
>>
>>> *From: *Marco Ocana <marco.ocana at balboasystems.com 
>>> <mailto:marco.ocana at balboasystems.com>>
>>> *Date: *October 13, 2010 8:22:46 AM PDT
>>> *To: *"Natasha F. Noy" <noy at stanford.edu <mailto:noy at stanford.edu>>, 
>>> Trish Whetzel <whetzel at stanford.edu <mailto:whetzel at stanford.edu>>
>>> *Cc: *Paolo Ciccarese <paolo.ciccarese at gmail.com 
>>> <mailto:paolo.ciccarese at gmail.com>>
>>> *Subject: **Proxy error when calling text mining service*
>>>
>>> Hi Natasha and Trish,
>>>
>>> On one of my test cases I am consistently getting an error from the 
>>> text mining service. I was able to reproduce it using the html form 
>>> at http://rest.bioontology.org/test_oba.html. Below is the text from 
>>> the returned web page:
>>>
>>>
>>>   Proxy Error
>>>
>>> The proxy server received an invalid response from an upstream server.
>>> The proxy server could not handle the request /POST /obs/annotator 
>>> <http://rest.bioontology.org/obs/annotator>/.
>>> Reason: *Error reading from remote server*
>>> *
>>> *
>>> To reproduce from the online form, below are the steps:
>>>
>>> 1. For ontologiesToKeepInResult:
>>> 42584,42801,3083,13356,39183,40009,42672,44158,42983,13352,40724,40726,36430,44005,44006,44007,39976,42277,42571,40791,44163,40133,3093,39579,39577,44169,42685,13342,42971,44133,42504,38631,42502,39966,40917,42728,40216,40688,44079,36417,44170,42925,44171,44172,44173,42920,35686,44174,44175,40546,39509,44178,42651,40872,42791,39508,13404,42860,39580,42794,40483,40117,36425,25148,42456,44182,42915,42455,42912,40802,44180,42453,42452,40223,44112,42093,42993,44113,40937,44084,44016,42525,42998,44011,39624,37911,42545,42646,8056,40764,40660,44057,42117,13397,44108,42948,29684,40245,39215,44109,39933,40865,42828,40652,39343,40651,38873,40653,42736,42737,39734,13386,42931,42932,40958,42330,44448,42835,42838,42837,40238,42834,39927,42764,42622,40832,39331,44038,40649,40648,3905,40646,42768,44036,42765,40643,40642,14391,42367,40261,39815,42820,39336,39544,39038,40844,42752,13323,42758,40984,4531,40985,42184,38801,38802,38550 
>>> ,42497,4525,38366,40358,42955,39804
>>>
>>> 2. For textToAnnotate, I copied and pasted the text in attached 
>>> file. Note: need to open in an editor that can read UTF-8 . The test 
>>> for this is that you can find:  Aβ    in the document.
>>>
>>> 3. For format, xml.
>>>
>>> If this is something you can fix at your end and/or have a 
>>> workaround, it would be greatly appreciated.
>>>
>>> Thanks
>>>
>>> Marco
>>>
>>>
>>>
>>
>>
>>
>>
>> _______________________________________________
>> bioontology-support mailing list
>> bioontology-support at lists.stanford.edu
>> https://mailman.stanford.edu/mailman/listinfo/bioontology-support
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.stanford.edu/pipermail/bioontology-support/attachments/20101018/27d70b3d/attachment.html>


More information about the bioontology-support mailing list