Re: Contributing DBSCAN algorithm for Mahout

[Grant Ingersoll <gs...@apache.org>]

On Sep 26, 2012, at 5:46 AM, Kakarala Rajendra wrote:

> Hi,
> 
>             As part of Data Mining course in my graduate school, our
> project team wants to contribute DBSCAN algorithm using
> map-reduce for Mahout.
> 
>              In regard to this please provide some inputs regarding
> 
>             1. Among several variants of DBSCAN algorithms proposed in
> several research papers, we have choosen QIDBSCAN.
>                Any idea of other DBSCAN algorithms that might be useful
> for the community if implemented.

Generally speaking, we are trying to focus on solidifying the code we have, but if you put up a well tested implementation, it will likely get committed at some point not too far off as it fits with our clustering libraries.


> 
>            2. Information regarding the coding conventions and standards
> to be used, inputs regarding how to make the code generic so
> that many applications can use.
> 

See the How To Contribute page on the Wiki.


>            Any other suggestions are also welcome.
> 
> 
> 
> 
> Thanks and Regards,
> Kakarala Rajendra,
> M.Tech First Year,
> Dept of Computer Science and Engineering,
> Indian Institute of Technology,
> Guwahati, India

--------------------------------------------
Grant Ingersoll
http://www.lucidworks.com