Deduplication with Nutch

nutch-site.xml

<property>
  <name>db.signature.class</name>
  <value>org.apache.nutch.crawl.TextProfileSignature</value>
  <description>The default implementation of a page signature. Signatures
  created with this implementation will be used for duplicate detection
  and removal.</description>
</property>

<property>
  <name>db.signature.text_profile.min_token_len</name>
  <value>2</value>
  <description>Minimum token length to be included in the signature.
  </description>
</property>

<property>
  <name>db.signature.text_profile.quant_rate</name>
  <value>0.01</value>
  <description>Profile frequencies will be rounded down to a multiple of
  QUANT = (int)(QUANT_RATE * maxFreq), where maxFreq is a maximum token
  frequency. If maxFreq > 1 then QUANT will be at least 2, which means that
  for longer texts tokens with frequency 1 will always be discarded.
  </description>
</property>