mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Kamesh <>
Subject creating mahout vector from record
Date Tue, 10 Jun 2014 17:23:01 GMT
Hi All,
 I am using Mahout first time and I want to use Canopy clustering
algorithm. I want to cluster similar records.

 Each record consists of both string and numeric values. To convert a
record into a Vector, I am assigning each record, a unique record id and
creating vector with that record id.
 When comparing distance between vectors, I am fetching corresponding
records for that record ids in the distance measure function and
calculating distance.

 Are there any approaches in Mahout to convert this kind of records into
mahout vectors. If so, please point me.


  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message