XML-based Natural Language Generation

Extra Assignment: Estonian Generation

This assignment is required to obtain a 2nd course credit.

• First, make a small "corpus" of Estonian examples
  • Make the examples like the English liter.examples
  • Write the Estonian examples in a file ee.liter.examples
• Build a complete Estonian NLG pipeline.
  • Ensure the Ant build file build2.xml is in XMLGen
  • Ensure NLGPipeline.java is in XMLGen/src
  • Ensure liter.xml is in XMLGen/data
• For the text planning stages, you can keep the same stylesheets for both English and Estonian, or you can write new ones.
  • Ensure your version of liter.content.xslt is in XMLGen/xslt
    Run "ant -f build2.xml nlgpipecontent" to do content determination only
  • Ensure your version of liter.discourse.xslt is in XMLGen/xslt
    Run "ant -f build2.xml nlgpipediscourse" to do content determination and discourse planning.
    This should show the text plan
• Write new Estonian stylesheets for the microplanning stages. You can keep "null" aggregation.
  • Ensure your Estonian version of liter.lexical.xslt is in XMLGen/xslt
    Run "ant -f build2.xml nlgpipelexical" to do lexicalization after text planning
  • Ensure liter.aggreg.xslt is in XMLGen/xslt
    Run "ant -f build2.xml nlgpipeaggreg" to do lexicalization and (null) aggregation after text planning
  • Ensure your Estonian version of liter.refexp.xslt is in XMLGen/xslt
    Run "ant -f build2.xml nlgpiperefexp" to do text planning, lexicalization, aggregation and referring expressions.
    This should show the text specification
• Write a new Estonian stylesheet for the surface realization stage.
  • Ensure your Estonian version of liter.realize.xslt is in XMLGen/xslt
    Run "ant -f build2.xml nlgpiperealize" to do text planning, microplanning and surface realization.
    This should show the Estonian text generated by the complete NLG pipeline
• Write a brief report about making the Estonian generation
  • Describe the problems and how you solved them
  • The report should be only one page
  • Write it in a plain text file ee.report.txt
• Send me the following files
  • Your Estonian "corpus" file ee.liter.examples
  • Your stylesheets liter.content.xslt, liter.discourse.xslt,
    liter.lexical.xslt, liter.aggreg.xslt, liter.refexp.xslt,
    liter.realize.xslt
  • Your report ee.report.txt