[R] question about adaboost.

[Uwe Ligges]

Cecilia Lezama wrote:
> Thanks for your quickly answer. I feel embarrassed but I didn't 
> understand it.
> 
> 1 - diag(table) / sum(table)        Which diagonal? Both of them?
> 
> Could you explain it to me with a practical example?

Whoops, this meant to be

1 - sum(diag(table)) / sum(table)

i.e. 1 - (138+2)/(138+2+9+1)

Uwe Ligges


> CONFUSION MATRIX
> 
>                            Observed Class
> 
> Predicted Class             A          P
> 
>              A                   138       9
> 
>              P                     1         2
> 
> 
> PROBABILITY CONFUSION MATRIX
> 
>                                Observed Class
> 
> Predicted Class                 A           P
> 
>              A       0.920000000  0.060000000
> 
>              P       0.006666667   0.013333333
> 
> 
> Many, many thanks!
> 
> 
> ----- Original Message ----- From: "Uwe Ligges" 
> <ligges at statistik.tu-dortmund.de>
> To: "Cecilia Lezama" <checha at netgate.com.uy>
> Cc: <r-help at r-project.org>
> Sent: Tuesday, April 28, 2009 9:42 AM
> Subject: Re: [R] question about adaboost.
> 
> 
>>
>>
>> Cecilia Lezama wrote:
>>> Hello,
>>> I would like to know how to obtain the misclassification error when 
>>> performing a boosting analisis with ADABAG package?
>>> With:
>>>> prop.table(Tesis.boostcv$confusion)
>>>
>>> I obtain the confusion matrix, but not the overall missclassification 
>>> error.
>>
>> Well, the misclassification error is
>>
>>   1 - diag(table) / sum(table)
>>
>> Uwe Ligges
>>
>>
>>
>>> Thanks in advance,
>>>
>>>
>>> BSc. Cecilia Lezama
>>> Facultad de Ciencias - UDELAR
>>> Montevideo - Uruguay.
>>> [[alternative HTML version deleted]]
>>>
>>> ______________________________________________
>>> R-help at r-project.org mailing list
>>> https://stat.ethz.ch/mailman/listinfo/r-help
>>> PLEASE do read the posting guide 
>>> http://www.R-project.org/posting-guide.html
>>> and provide commented, minimal, self-contained, reproducible code.
>>
> 
>