GATE and UIMA in Language Technology Teaching Graham Wilcock - PowerPoint PPT Presentation

GATE and UIMA in Language Technology Teaching Graham Wilcock University of Helsinki graham.wilcock@helsinki.fi

Outline • Current course materials • Shakespeare’s Sonnets • GATE & ANNIE • New course materials • UIMA & OpenNLP • UIMA & Stanford NLP • Appendix • Eclipse & Stanford Eclipse Graham Wilcock UIMA workshop, GLDV-2007 2

IBM: Tidwell XSLT tutorials Graham Wilcock UIMA workshop, GLDV-2007 3

Gutenberg: Sonnets corpus Graham Wilcock UIMA workshop, GLDV-2007 5

Outline • Current course materials • Shakespeare’s Sonnets • GATE & ANNIE • New course materials • UIMA & OpenNLP • UIMA & Stanford NLP • Appendix • Eclipse & Stanford Eclipse Graham Wilcock UIMA workshop, GLDV-2007 6

GATE & ANNIE • Start with ANNIE • Ready-to-run NLP tools • ANNIE NE Recognizer • ANNIE POS Tagger • Add JAPE annotations • Students write NP, PP rules Graham Wilcock UIMA workshop, GLDV-2007 7

GATE & ANNIE • Start with ANNIE • Ready-to-run NLP tools • ANNIE NE Recognizer • ANNIE POS Tagger • Add JAPE annotations • Students write NP, PP rules Graham Wilcock UIMA workshop, GLDV-2007 9

GATE & ANNIE • Start with ANNIE • Ready-to-run NLP tools • ANNIE NE Recognizer • ANNIE POS Tagger • Add JAPE annotations • Students write NP, PP rules Graham Wilcock UIMA workshop, GLDV-2007 11

Outline • Current course materials • Shakespeare’s Sonnets • GATE & ANNIE • New course materials • UIMA & OpenNLP • UIMA & Stanford NLP • Appendix • Eclipse & Stanford Eclipse Graham Wilcock UIMA workshop, GLDV-2007 13

UIMA & OpenNLP • OpenNLP tools • First, run from command line • Install in UIMA (assignment) • UIMA & OpenNLP • OpenNLP POS Tagger • OpenNLP NE Recognizer Graham Wilcock UIMA workshop, GLDV-2007 14

UIMA & OpenNLP • OpenNLP tools • First, run from command line • Install in UIMA (assignment) • UIMA & OpenNLP • OpenNLP POS Tagger • OpenNLP NE Recognizer Graham Wilcock UIMA workshop, GLDV-2007 15

UIMA & OpenNLP • OpenNLP tools • First, run from command line • Install in UIMA (assignment) • UIMA & OpenNLP • OpenNLP POS Tagger • OpenNLP NE Recognizer Graham Wilcock UIMA workshop, GLDV-2007 17

UIMA & OpenNLP • OpenNLP tools • First, run from command line • Install in UIMA (assignment) • UIMA & OpenNLP • OpenNLP POS Tagger • OpenNLP NE Recognizer • Add Java annotators • Students write NP, PP annotators? Graham Wilcock UIMA workshop, GLDV-2007 19

Outline • Current course materials • Shakespeare’s Sonnets • GATE & ANNIE • New course materials • UIMA & OpenNLP • UIMA & Stanford NLP • Appendix • Eclipse & Stanford Eclipse Graham Wilcock UIMA workshop, GLDV-2007 20

Stanford NLP Group • Stanford NLP tools • Stanford POS Tagger • Stanford NE Recognizer • Stanford Parser Graham Wilcock UIMA workshop, GLDV-2007 21

UIMA & Stanford NLP • Stanford NE Recognizer • Use Stanford NER-GUI • UIMA & Stanford NER • Install in UIMA (assignment) • UIMA wrapper by F. Laws Graham Wilcock UIMA workshop, GLDV-2007 22

UIMA & Stanford NLP • Stanford NE Recognizer • Use Stanford NER-GUI • UIMA & Stanford NER • Install in UIMA (assignment) • UIMA wrapper by F. Laws Graham Wilcock UIMA workshop, GLDV-2007 24

Outline • Current course materials • Shakespeare’s Sonnets • GATE & ANNIE • New course materials • UIMA & OpenNLP • UIMA & Stanford NLP • Appendix • Eclipse & Stanford Eclipse Graham Wilcock UIMA workshop, GLDV-2007 26

Eclipse & Stanford Eclipse • jEdit vs. Eclipse • Students currently use jEdit • Eclipse learning curve is harder • Stanford Eclipse • Stanford CS: simplified Eclipse • Karel the Robot Learns Java • Starter projects for assignments Graham Wilcock UIMA workshop, GLDV-2007 27

Starter projects Graham Wilcock UIMA workshop, GLDV-2007 29

Summary • Current course materials • Shakespeare’s Sonnets • GATE & ANNIE • New course materials • UIMA & OpenNLP • UIMA & Stanford NLP • Appendix • Eclipse & Stanford Eclipse Graham Wilcock UIMA workshop, GLDV-2007 33