Integration and Conclusion

Part of the Lecture Notes in Computer Science book series (LNCS, volume 1744)


Beginning with the discussion of the lexical semantics of degree expressions in Chapter 3, a working strategy has been pursued that split the degree information extraction task into the three subtasks of defining comparison relations (and inferences on them), computing comparison objects, and computing comparison classes. Interactions between the two comparison paradigms and the representation and inferencing scheme, as well as differences and commonalities between relations and intercorrelations, have mostly been neglected, so far. Approaching such an integration of methods in a unified view, these loose ends are tied up in this chapter. Nevertheless, the purpose of integration raises new research issues, which can barely be touched upon in this chapter. The reason is that the integration requires heuristic combinations that call for an all-encompassing empirical evaluation of the integrated modules — a enormous task which could not be tackled in this book. After having outlined this new research agenda, I point out some desiderata of general concern to the degree information extraction task, before I finally conclude.


Inkjet Printer Comparison Relation Class Norm Comparison Class Linguistic Knowledge 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 1999

Personalised recommendations