Canopy clustering over strings and the input format and hadoop implementation

See more:

Java

I want to do canopy clustering over strings to reduce the distance and the measures. But I not having any idea how to do canopy clustering over set of strings.
When I searched I got the Apache hadoop implementation of text clustering. But in that they said the input format should be sequential vector file in which the input should vector readable format.


I have a column of strings and how to change this into sequential file and vector file in java and how to use hadoop canopy clustering efficiently.


example of one column words :

quickly
need
close
this?
father-in-law
relatives
come
location?
little
specific
'where
exactly
chennai-bangalore
road?',
far
road?
state
right
locality
in?
post
message
brahmma
weeks
max


help me thanks