'Re: [Wekalist] LOF and removing outliers form data that are consisted of [nominal attributes and nom'

[prev in list] [next in list] [prev in thread] [next in thread] 

List: wekalist
Subject: Re: [Wekalist] LOF and removing outliers form data that are consisted of [nominal attributes and nom
From: Leonardo Lion <llionname () gmail ! com>
Date: 2014-11-26 13:20:40
Message-ID: CAH8X3+GuXu10VTkNytV5AneTE4zyk3UPLfG4UShCc3KUcq0M_g () mail ! gmail ! com
[Download RAW message or body]

[Attachment #2 (multipart/alternative)]

Hi Martin,

You can remove outliers through applying classification method. After that,
use RemoveMisclassified filter to delete the instences that are
misclassified.

Regards,
Leonardo
On Nov 25, 2014 8:02 PM, "Martin L" <martinlion80@gmail.com> wrote:

> Hi all,
>
>
>
> In order to remove outliers from the dataset, I used
> weka.filters.unsupervised.attribute. LOF which I think that it just can be
> used to compute the outliers by calculating some main statistical metrics
> (min, max, mean, SD). After pressing the button "Apply" in Weka to apply
> LOF on the datatset, this filter generates an attribute that contains clear
> information about the existed outliers. Now I have to remove this
> *attribute* since I had the outliers' information from the one hand, and
> I have to remove those *outliers* that were mentioned by LOF from the
> other hand.
>
>
> Therefore,
>
> a)      If LOF finds 100 outliers, what is the best way (filter) to
> remove them in    particularly since it does not have an option to delete
> them (I as I believe)?
> b)      With the case of [nominal attributes and nominal class], what is
> the best Weka's filer to remove the outliers?
>
> Thanks.
> Martin
>
>
>
> _______________________________________________
> Wekalist mailing list
> Send posts to: Wekalist@list.waikato.ac.nz
> List info and subscription status:
> http://list.waikato.ac.nz/mailman/listinfo/wekalist
> List etiquette:
> http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
>
>

[Attachment #5 (text/html)]

Hi Martin,
You can remove outliers through applying classification method. After that, use \
RemoveMisclassified filter to delete the instences that are misclassified. \
Regards, Leonardo
<div class="gmail_quote">On Nov 25, 2014 8:02 PM, &quot;Martin L&quot; &lt;<a \
href="mailto:martinlion80@gmail.com">martinlion80@gmail.com</a>&gt; wrote: <blockquote class="gmail_quote" style="margin:0 0 0 \
.8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr"><div><div>Hi \
all, In \
order to remove outliers from the dataset, I used \
weka.filters.unsupervised.attribute. LOF which I think that it just can be used to \
compute the outliers by calculating some main statistical metrics (min, max, mean, \
SD). After pressing the button "Apply" in Weka to apply LOF on the datatset, this \
filter generates an attribute that contains clear information about the existed \
outliers. Now I have to remove this attribute since I had the outliers' \
information from the one hand, and I have to remove those outliers that were \
mentioned by LOF from the other hand.

<p class="MsoNormal" style="margin-bottom:0.0001pt;line-height:normal"><span \
style="font-size:12pt;font-family:&quot;Times New \
Roman&quot;,&quot;serif&quot;"><br></span></p><p class="MsoNormal" \
style="margin-bottom:0.0001pt;line-height:normal"><span \
style="font-size:12pt;font-family:&quot;Times New \
Roman&quot;,&quot;serif&quot;">Therefore, </span></p>

a) If \
LOF finds 100 outliers, what is the best way (filter) to remove them in \
particularly since it does not have an option to delete them (I as I believe)? \

<span style="font-size:12pt;font-family:&quot;Times New \
Roman&quot;,&quot;serif&quot;"><span>b)<span style="font:7pt &quot;Times New \
Roman&quot;">           </span></span></span><span dir="LTR"></span><span \
style="font-size:12pt;font-family:&quot;Times New Roman&quot;,&quot;serif&quot;">With \
the case of [nominal attributes and nominal class], what is the best Weka's filer
to remove the outliers?<br><br></span></div><span \
style="font-size:12pt;font-family:&quot;Times New \
Roman&quot;,&quot;serif&quot;">Thanks.<br></span></div><span \
style="font-size:12pt;font-family:&quot;Times New \
Roman&quot;,&quot;serif&quot;">Martin<br></span><span \
style="font-size:12pt;font-family:&quot;Times New Roman&quot;,&quot;serif&quot;"> \
</span><p style="margin-bottom:0.0001pt;line-height:normal"><br></p></div> \
<br>_______________________________________________<br> Wekalist mailing list<br>
Send posts to: <a href="mailto:Wekalist@list.waikato.ac.nz">Wekalist@list.waikato.ac.nz</a><br>
 List info and subscription status: <a \
href="http://list.waikato.ac.nz/mailman/listinfo/wekalist" \
target="_blank">http://list.waikato.ac.nz/mailman/listinfo/wekalist</a><br> List \
etiquette: <a href="http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html" \
target="_blank">http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html</a><br>
 <br></blockquote></div>

_______________________________________________
Wekalist mailing list
Send posts to: Wekalist@list.waikato.ac.nz
List info and subscription status: http://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html

[prev in list] [next in list] [prev in thread] [next in thread] 