News for hackers. Stuff that matters.

A comment on The Berkeley Document Summarizer: Learning-Based, Single-Document Summarization. Parent comment

Re:The Berkeley Document Summarizer: Learning-Based, Single-... by ganeshkrishnan on Sunday October 02, 2016 @07:31AM (#12621898)
We work with NLP and Java. Stanford NLP and apache Spark both run on Java/Scala and so does hadoop for HDFS. Why the hate against Java?
[ Parent | on HN ]
Re:The Berkeley Document Summarizer: Learning-Based, Single-... by amelius on Sunday October 02, 2016 @12:56PM (#12623198)
Startup time and memory footprint?
[ Parent | on HN ]
Re:The Berkeley Document Summarizer: Learning-Based, Single-... by ganeshkrishnan on Friday October 14, 2016 @12:44AM (#12706184)
Its hdfs so distributed across machines. Startup time is not a problem for us since the service seldom shutsoff
[ Parent | on HN ]