http://werner.ira.uka.de/ISL.speech.publications.html
--------------------------------------------------------------------------
FEASPAR - A FEATURE STRUCTURE PARSER LEARNING TO PARSE SPONTANEOUS SPEECH
(120 pages)
Finn Dag Buo
Ph.D thesis
University of Karlsruhe
Abstract
Traditionally, automatic natural language parsing and translation have been
performed with various symbolic approaches. Many of these have the advantage
of a highly specific output formalism, allowing fine-grained parse analyses
and, therefore, very precise translations. Within the last decade, statistical,
and connectionist techniques have been proposed to learn the parsing task in
order to avoid the tedious manual modeling of grammar and malformation. How to
learn a detailed output representation and how to learn to parse robustly even
ill-formed input, has until now remained an open question.
This thesis provides an answer to this question by presenting a connectionist
parser that needs a small corpus and a minimum of hand modeling, that learns,
and that is robust towards spontaneous speech and speech recognizer effects.
The parser delivers feature structure parses, and has a performance comparable
to a good hand modeled unification based parser.
The connectionist parser FeasPar consists of several neural networks and
a Consistency Checking Search. The number of, architecture of, and other
parameters of the neural networks are automatically derived from the training
data. The search finds the combination of the neural net outputs that produces
the most probable consistent analysis.
To demonstrate learnability and robustness, FeasPar is trained with
transcribed sentences from the English Spontaneous Scheduling Task and
evaluated for network, overall parse, and translation performance, with
transcribed and speech data. The latter contains speech recognition errors.
FeasPar requires only minor human effort and performs better or comparable
to a good symbolic parser developed with a 2 year, human expert effort.
A key result is obtained by using speech data to evaluate the JANUS
speech-to-speech translation system with different parsers. With FeasPar,
acceptable translation performance is 60.5 %, versus 60.8 % with a GLR* parser.
FeasPar requires two weeks of human labor to prepare the lexicon and 600
sentences of training data, whereas the GLR* parser required significant
human expert grammar modeling.
Presented in this thesis are the Chunk'n'Label Principle, showing how to divide
the entire parsing tasks into several small tasks performed by neural networks,
as well as the FeasPar architecture, and various methods for network performance
improvement. Further, a knowledge analysis and two methods for improving the
overall parsing performance are presented. Several evaluations and comparisons
with a GLR* parser, producing exactly the same output formalism, illustrate
FeasPar's advantages.
================================================================================
Finn Dag Buo
SAP AG
Germany
finn.buoe@sap-ag.de
================================================================================