[ANN] acts_as_classifiable plugin


#1

Hello,
I am pleased to announce the availability of the plugin
acts_as_classifier
which allows using the ‘classifier’ gem in a Rails application.

This plugin can be downloaded from

http://opensvn.csie.org/sksinghi/acts_as_classifiable/

This plugin is useful in scenarios where you want to distinguish between
spam
or non-spam comments, Or maybe you want to track the preferences of each
user
and then based on that make suggestions to them, etc.

More information + example on this can be found at:

http://cuttingtheredtape.blogspot.com/2006/05/actsasclassifiable.html

Any issues/questions let me know.


Surendra S.
http://ssinghi.kreeti.com, http://www.kreeti.com
Read my blog at: http://cuttingtheredtape.blogspot.com/
,----
| “War is Peace! Freedom is Slavery! Ignorance is Strength!”
| – Orwell, 1984, 1948
`----


#2

Published at http://agilewebdevelopment.com/plugins/
acts_as_classifiable :slight_smile:


Benjamin C.
http://www.bencurtis.com/
http://www.tesly.com/ – Collaborative test case management
http://www.agilewebdevelopment.com/ – Resources for the Rails community


#3

That’s great Surendra.

(having read your posted information) For those of us not familiar
with Bayesian classifiers, I’m don’t understand how it would be used
to classify.

You mention training - are we required to implement methods to do the
classification?

I definetly have some requirements for such a gem, but am unsure how
I’d go about actually using it.

Thanx again.

Jodi


#4

grin. thanx Peter (i’m the one to usually suggest obvious references.
lol) . I think that wp entry gave me enough info to start.

ie. the specific application of training and initial classification
are the keys to usage.

cheers,
J


#5

(having read your posted information) For those of us not familiar
with Bayesian classifiers, I’m don’t understand how it would be
used to classify

http://en.wikipedia.org/wiki/Bayesian_filter

Best regards

Peter De Berdt


#6

Jodi S. removed_email_address@domain.invalid writes:

grin. thanx Peter (i’m the one to usually suggest obvious references.
lol) . I think that wp entry gave me enough info to start.

ie. the specific application of training and initial classification
are the keys to usage.

The current Bayesian classifier gem can do with some sprucing up[1].
Right
now, it doesn’t uses prior probabilities, no classification threshold,
cost-sensitive learning, etc.

But, it should be good for general usage. Also, tuning it do better will
require some experience with the domain where you are using it.

[1]I posted a message on the classifier gem forum, but haven’t got a
reply so
far, maybe I should try sending them a personal email.

Cheers.


Surendra S.
http://ssinghi.kreeti.com, http://www.kreeti.com
Read my blog at: http://cuttingtheredtape.blogspot.com/
,----
| “War is Peace! Freedom is Slavery! Ignorance is Strength!”
| – Orwell, 1984, 1948
`----