To Do List

• Segmentation/Tagging in one swoop
  • What has been done?
    • Hanna Wallach CRF tutorial
    • First CRF tutorial
    • Chinese segmentation/tagging combo papers.
  • Find where Lab is
    • Peter's state of affairs POS Tagging
    • Morph List??
    • Run morph tagger, perceptron tagger?

For experiments,

• Describe succinctly
• Reason
• Certain about method

Changes to StatNLP

• Turn FeatureExtractor into an abstract class that handles the if(Labeled) casting issues in all FeatureExtractors
• FeatureExtractor Interface: Type (maybe Datum maybe not) → Datum, Readers output a Type (for now we can stick with Collection of datums)
  • Another possibility is structured datum is a map which will change interfaces but may fit better
    • Generic feature extractor could extract features from the text (of email, doc, forum, etc.)
• Models take Datums (exceptions may exist like NetflixMovieModel but not in code base)
• pipelines?
• what is the role of vectors in the feature pipeline?