nlp_playtime_1

nlp_playtime.rb

# need jruby
# need stanford-corenlp in a directory somewhere
# untar it, cd into it

# irb innvocation for more memory happy:
# jruby -J-Xmn512m -J-Xms2048m -J-Xmx2048m -J-server -S irb

require 'java'
require 'stanford-corenlp-2011-09-16.jar'
require 'stanford-corenlp-2011-09-14-models.jar'
require 'xom.jar'
require 'joda-time.jar'

java_import "edu.stanford.nlp.pipeline.StanfordCoreNLP"

nlp = StanfordCoreNLP.new
# => NativeException: java.lang.NoClassDefFoundError: Could not initialize class
# edu.stanford.nlp.time.SUTime


java_import "edu.stanford.nlp.process.DocumentPreprocessor"
dp = DocumentPreprocessor.new("input.txt")
dp.entries.to_s
# "[Stanford, University, is, located, in, California, .][It, is, a, great, university, .]"

words = dp.entries.collect {|e| e.collect {|word| word.to_s} }

words.index {|sentence| sentence.include? "great"}
# => 1
words.index {|sentence| sentence.include? "in"}
# => 0