KudoZ home » Italian to English » Mathematics & Statistics

90% complementare

English translation: the rest of the sample

Advertisement

Login or register (free and only takes a few minutes) to participate in this question.

You will also have access to many other tools and opportunities designed for those who have language-related jobs
(or are passionate about them). Participation is free and the site has a strict confidentiality policy.
10:38 Oct 23, 2007
Italian to English translations [PRO]
Science - Mathematics & Statistics / statistical analysis
Italian term or phrase: 90% complementare
La capacità predittiva del modello è stata misurata tramite il metodo di validazione interna “cross validation”... stimando il modello sul 90% complementare di ogni gruppo e riapplicandolo sul gruppo selezionato.

Can any statistics whizz help me please? There are no hits at all with google and having a look for likely translations hasn't come up with much.
Marie-Hélène Hayles
Local time: 14:57
English translation:the rest of the sample
Explanation:
or "the remaining 90% of the original sample", maybe; or even something like "...calculating model estimates, for each group, on its complement (90% of the whole sample)...."

I'm sure you can find a better way to say it in English, but this might be a starting point...

Reading the Italian text at

( http://www.iss.it/binary/publ/publi/404393.1111738552.pdf ),

I understand that for the purpose of cross-validation they divided (randomly) the sample in 10 sub-groups and then, for each sub- group, pooled the remaining 9 sub-groups into a larger one, comprised (obviously) of 90% of the individual items of the original sample - and then went on to perform whatever calculations they needed to.

I think that part of the problem stems from overuse, in the Italian text, of the word "gruppo"; and also from the incorrect, albeit widely used, expression "stimare il modello" (you can estimate parameters, not models: in fact, statistical models are used to estimate parameters...)
Selected response from:

Alfredo Tutino
Local time: 14:57
Grading comment
I've selected Alfredo's answer over Bruna's because I had a pressing deadline and this really helped me understand what the authors were saying. I finally translated it as "comparing each sub-group against the pool of the other 9 groups"; Bruna's suggestion would have been more professional but arrived too late to be incorporated!
4 KudoZ points were awarded for this answer

Advertisement


Summary of answers provided
3complementary to 90%
Gian
390% of the sample was used as a complement
texjax DDS PhD
2the rest of the sampleAlfredo Tutino


  

Answers


17 mins   confidence: Answerer confidence 3/5Answerer confidence 3/5
complementary to 90%


Explanation:
Characterization of the mRNA of influenza virus.- [ Traduci questa pagina ]By the following criteria, the complementary RNA had properties normally ... whereas poly(A)-minus RNA contained sequences complementary to 90 to 100% of ...
www.pubmedcentral.nih.gov/articlerender.fcgi?artid=355752 - Pagine simili


Gian
Italy
Local time: 14:57
Works in field
Native speaker of: Native in ItalianItalian
PRO pts in category: 34

Peer comments on this answer (and responses from the answerer)
neutral  Alexander Chisholm: the example provided has nothing to do with the context here - complimentary here relates to the degree of identity or homology between two sequences (RNA sequence in this case)
28 mins
  -> a me interessa dire come si usa complementary; il complemnto a... significa il resto e per i gradi di un triangolo rett. è il complemento a 90°.

neutral  Alfredo Tutino: with Alexander
1 hr
Login to enter a peer comment (or grade)

2 hrs   confidence: Answerer confidence 2/5Answerer confidence 2/5
the rest of the sample


Explanation:
or "the remaining 90% of the original sample", maybe; or even something like "...calculating model estimates, for each group, on its complement (90% of the whole sample)...."

I'm sure you can find a better way to say it in English, but this might be a starting point...

Reading the Italian text at

( http://www.iss.it/binary/publ/publi/404393.1111738552.pdf ),

I understand that for the purpose of cross-validation they divided (randomly) the sample in 10 sub-groups and then, for each sub- group, pooled the remaining 9 sub-groups into a larger one, comprised (obviously) of 90% of the individual items of the original sample - and then went on to perform whatever calculations they needed to.

I think that part of the problem stems from overuse, in the Italian text, of the word "gruppo"; and also from the incorrect, albeit widely used, expression "stimare il modello" (you can estimate parameters, not models: in fact, statistical models are used to estimate parameters...)

Alfredo Tutino
Local time: 14:57
Works in field
Native speaker of: Native in ItalianItalian
PRO pts in category: 12
Grading comment
I've selected Alfredo's answer over Bruna's because I had a pressing deadline and this really helped me understand what the authors were saying. I finally translated it as "comparing each sub-group against the pool of the other 9 groups"; Bruna's suggestion would have been more professional but arrived too late to be incorporated!
Notes to answerer
Asker: That's a great explanation Alfredo - thanks very much.

Login to enter a peer comment (or grade)

7 hrs   confidence: Answerer confidence 3/5Answerer confidence 3/5
90% of the sample was used as a complement


Explanation:
Hi Marie-Hélène, you might want to take a look at this

We used a cross-validation process to determine the optimal number of antibodies in the final combined classifier for all three methods. For the boosting methods, a 10-fold cross-validation step is applied, where 90% of the samples are used as training set to define a best model for classification whereas 10% of the samples are reserved as testing set to determine the error rate of the model. This process is repeated 10 times, each time using a different group of 90% for classification and the cross-validation error is the average of the 10 error rates. The classifier is considered final when the further addition of an antibody will not further decrease the cross-validation error.

http://cancerres.aacrjournals.org/cgi/content/full/65/23/111...

--------------------------------------------------
Note added at 7 hrs (2007-10-23 17:52:34 GMT)
--------------------------------------------------

Complement: the opposite of an event; the set of outcomes not belonging to an event.

--------------------------------------------------
Note added at 8 hrs (2007-10-23 18:41:29 GMT)
--------------------------------------------------

You can also say "used as training set" as in the provided example.
Anyway, I believe complement is the correct statistical term here.



texjax DDS PhD
Local time: 08:57
Specializes in field
Native speaker of: Native in ItalianItalian
PRO pts in category: 40
Notes to answerer
Asker: Thanks Bruna - I think that essentially confirms what Alfredo suggests, putting it in more technical language.

Login to enter a peer comment (or grade)




Return to KudoZ list


KudoZ™ translation help
The KudoZ network provides a framework for translators and others to assist each other with translations or explanations of terms and short phrases.



See also:



Term search
  • All of ProZ.com
  • Term search
  • Jobs
  • Forums
  • Multiple search