[NUTCH-2136] Implement a different version of Naive Bayes Parse Filter

• Type: Improvement
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 1.10
• Component/s: parser
• Assignee: Unassigned
• Reporter: Asitang Mishra
• Created: 12/Oct/15 03:24
• Updated: 13/Mar/24 14:50
• Resolved: 12/Oct/15 18:04

Description

There has been many dependency issues with the first implementation of Naive Bayes Parse Filter. The major dependencies were Mahout and Lucene. There was also the issue where the training process failed in the distributed mode due to the fact that a nested hadoop job was unable to run on the cluster.
To remove all these issues and make the filter be able to run in a distributed environment I am going to implement my own version of Naive Bayes without any dependency on any machine learning libraries.

Issue Links

• is related to: NUTCH-2137 add changes.txt and ALV2 headers to the Naive Bayes Parse Filter