About | Study | Members | Projects | Publications | Events | Search | Internals | Contact & Imprint
Course title: Advanced Topics in Weighted Tree Automata
Lecturer: Prof. Heiko Vogler
Module: MCL-NLPA
Status: advanced unit
SWS (lecture/tutorial/practical): 4/2/0
Examination method: to be announced
In many areas of Computer Science trees are used to structure data and to represent hierarchical dependencies between parts of data; for instance, the structure of a sentence of a natural language is represented as a tree (parse tree). Hence, it is of general interest to provide algorithms and machines which decide whether trees have certain properties, rank trees by associating a weight with each of them, or transform trees into other trees. Taking up the natural language scenario, it is e.g. important to know the probability that a sentence is well-structured with respect to some grammatical rules, or what the possible translations of a sentence of one language into another language are.
Weighted tree automata are appropriate structures for such algorithms. In this lecture we will introduce their basic definitions and properties on a theoretical level, however keeping the application to natural language processing in mind.