giraph-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Georgos Siganos <siga...@gmail.com>
Subject Re: kmeans.
Date Tue, 22 Apr 2014 06:41:11 GMT
Hi Suijian,
You can find an initial implementation of k-means in the following url. It
will be part of okapi at some point in the future.
https://github.com/vasia/okapi/blob/kmeans_aggregator_per_center/src/main/java/ml/grafos/okapi/kmeans
Cheers,
Georgos


2014-04-22 0:36 GMT+03:00 Suijian Zhou <suijian.zhou@gmail.com>:

> Thanks Mirko, will try it.
>
> Best Regards,
> Suijian
>
>
>
>
>
> 2014-04-21 11:12 GMT-05:00 Mirko Kämpf <mirko.kaempf@cloudera.com>:
>
> Hi Suijian,
>>
>> in Grafos.ml (which builds on Giraph as well) you find:
>>
>>  Bayesian Personalized Ranking (BPR)
>>
>>    -  Overview <http://grafos.ml/okapi.html#tab13>
>>    -  Details <http://grafos.ml/okapi.html#tab14>
>>    -  How to Run <http://grafos.ml/okapi.html#tab15>
>>    -  References <http://grafos.ml/okapi.html#tab16>
>>
>> Bayesian Personalized Ranking from implicit feedback (BPR), the optimizes
>> Area Under the Curve (AUC) in a Collaborative Filtering setting. The
>> implementation in Okapi is a port of the myMediaLite implementation to the
>> giraph framework.
>>
>>
>>
>> k-Means clustering is also available in Mahout:
>>
>>       https://mahout.apache.org/users/clustering/k-means-clustering.html
>>
>>
>>
>>
>>
>>
>>
>>
>> and in GraphLab you find:
>>
>>
>> KMeans++
>>
>> The kmeans program implements the KMeans++<http://en.wikipedia.org/wiki/K-means%2B%2B>
algorithm
>> described by
>>
>> Arthur, D. and Vassilvitskii, S. (2007). "k-means++: the advantages of
>> careful seeding". Proceedings of the eighteenth annual ACM-SIAM symposium
>> on Discrete algorithms. pp. 1027–1035.
>>
>>       http://docs.graphlab.org/clustering.html
>>
>>
>>
>>
>> On Mon, Apr 21, 2014 at 6:04 PM, Suijian Zhou <suijian.zhou@gmail.com>wrote:
>>
>>> Hi,
>>>   Is there any example implementations of graph classification
>>> algorithms in giraph? E.g, k-means, Bayes etc?
>>>
>>>   Best Regards,
>>>   Suijian
>>>
>>>
>>
>>
>> --
>> --
>> Mirko Kämpf
>> mirko@cloudera.com
>>
>>
>

Mime
View raw message