1. Semantic annotation of a dialog corpus Silvie Cinková Institute of Formal and Applied Linguistics Charles University in Prague, Czech Republic COMPANIONS (www.companions-project.org)www.companions-project.org European Commission Sixth Framework Programme Information Society Technologies Integrated Project IST-34434

2. Data for machine learning audio-synchronized transcription linguistic annotation –Charles University (Czech Republic) –Napier University (Edinburgh, UK) –University of Sheffield (UK) –Oxford University (UK)

3. Functional Generative Description formal language description Prague structuralism + computational ling. since 1960's stratifies language –phonology –morphology –surface syntax –underlying syntax (tectogrammatics) transition between syntax and semantics a "poor men's interlingua"

4. Dependency  constituency syntax VP NP PP on the left ? thatJessIs NP

5. Tectogrammatical representation "Underlying syntax" linguistic meaning syntactic and semantic relations parent-child node(s) valency ellipsis restoration coreference across sentence boundaries information structure (TFA) synonymous function  identical representation

6. Tectogrammatical representation Is that Jess on the left?

7. Tectogrammatical representation ellipsis restoration coreference Yes it is, laughing.

8. Current... written –Prague Dependency Treebank Czech newspapers 800 k words manually LDC 2006 –Wall Street Journal in progress, 15% so far monolog reporting standard language spoken –dialogs real time interaction clause fragments exophora, deixis (syntax deviations) and challenges

9. Non-sentential utterances (NSU) phrases (NP, PP, ADVP, ADJP) –Me. –At 5 o'clock. –Blue. interjections –Mhm. –Oh, no! interjections attached to phrases –No, Billy. –Oh, sure. subordinate clause without main clause –If he goes with me. –Skiing. phrase combinations in coordination or apposition –With Mary in the morning or shopping at Tesco. –Or without.

10. Utterance-response pair "Who's that?" "Peggy." utterance U response NSU UPred UMods Functors (semantic labels)

11. Utterance-response pair Who's that? [Peggy.] Peggy.("That is Peggy").

12. Two students? Shopping with Mary. Coreferential predicate

13. Predicate with interjections No, Billy.Yes. Mhm.

14. NSUMods versus UMods attribute: response_type values: –overrules –bridging –wh-path –other form: reference (arrow) to antecedent node

15. Non-conflicting Modifier addition Yes [I brought the book]. [It will be] probably not [worth getting].

16. Overruling I'm at a little place called Ellenthorpe. Hellenthorpe.

17. Overruling by an identical modifier A: There are only two people in the class. B: Two people?

18. Bridging There are only two people in the class. Two students?

19. Bridging A: You lift the crane out, so this part will come up. B: The end?

20. Pronominal anaphora vs. overruling A: Peter should introduce Paul to Mary. B: Rather her to him.

21. Wh-path A: "Who's that?"B: "Peggy."

22. Wh-path - different functor matches up to the annotator we expect regular alternation patterns Where would you like to go tomorrow? Shopping with Mary.

23. Other A: He entered the largest room. B: Room 128? A: I don't know the number.

24. Summary U-NSU pairs NSU inherits the predicate of U (coreference) NSU inherits all modifiers of U NSU's own modifiers overrule the inherited –overrule –bridging –wh-path –other

25. References Raquel Fernández, Jonathan Ginzburg, and Shalom Lappin (2007): Classifying Non- Sentential Utterances in Dialogue: A Machine Learning Approach. Computational Linguistics, Volume 33, Nr. 3. MIT Press for the Association for Computational Linguistics Eva Hajičová (ed) (1995): Text-And- Inference-Based Approach to Question Answering, Prague, 1995