Linguistik-Klassifikation
Refine
Year of publication
Document Type
- Preprint (48)
- Conference Proceeding (19)
- Part of a Book (9)
- Article (8)
- Book (5)
- Working Paper (4)
Language
- English (93) (remove)
Has Fulltext
- yes (93)
Is part of the Bibliography
- no (93)
Keywords
- Computerlinguistik (33)
- Japanisch (15)
- Deutsch (13)
- Syntaktische Analyse (9)
- Maschinelle Übersetzung (8)
- Multicomponent Tree Adjoining Grammar (8)
- Lexicalized Tree Adjoining Grammar (5)
- Semantik (5)
- Satzanalyse (4)
- Transkription (4)
Institute
- Extern (80)
The definition of similarity between sentences is formulated on the levels of words, POS tags, and chunks (Abney 91; Abney 96). The evaluation of this approach shows that while precision and recall based on the PARSEVAL measures (Black et al. 91) do not reach state of the art Parsers yet (F1=87.19 on syntactic constituents, F1=77.78 including functionargument structure), the parser shows a very reliable performance where function-argument structure is concerned (F1=96.52). The lower F-scores are very often due to unattached constituents.
When a statistical parser is trained on one treebank, one usually tests it on another portion of the same treebank, partly due to the fact that a comparable annotation format is needed for testing. But the user of a parser may not be interested in parsing sentences from the same newspaper all over, or even wants syntactic annotations for a slightly different text type. Gildea (2001) for instance found that a parser trained on the WSJ portion of the Penn Treebank performs less well on the Brown corpus (the subset that is available in the PTB bracketing format) than a parser that has been trained only on the Brown corpus, although the latter one has only half as many sentences as the former. Additionally, a parser trained on both the WSJ and Brown corpora performs less well on the Brown corpus than on the WSJ one. This leads us to the following questions that we would like to address in this paper: - Is there a difference in usefulness of techniques that are used to improve parser performance between the same-corpus and the different-corpus case? - Are different types of parsers (rule-based and statistical) equally sensitive to corpus variation? To achieve this, we compared the quality of the parses of a hand-crafted constraint-based parser and a statistical PCFG-based parser that was trained on a treebank of German newspaper text.
Based on a detailed case study of parallel grammar development distributed across two sites, we review some of the requirements for regression testing in grammar engineering, summarize our approach to systematic competence and performance profiling, and discuss our experience with grammar development for a commercial application. If possible, the workshop presentation will be organized around a software demonstration.
The Deep Linguistic Processing with HPSG Initiative (DELH-IN) provides the infrastructure needed to produce open-source semantic transfer-based machine translation systems. We have made available a prototype Japanese-English machine translation system built from existing resources include parsers, generators, bidirectional grammars and a transfer engine.
In this paper we describe SOBA, a sub-component of the SmartWeb multi-modal dialog system. SOBA is a component for ontologybased information extraction from soccer web pages for automatic population of a knowledge base that can be used for domainspecific question answering. SOBA realizes a tight connection between the ontology, knowledge base and the information extraction component. The originality of SOBA is in the fact that it extracts information from heterogeneous sources such as tabular structures, text and image captions in a semantically integrated way. In particular, it stores extracted information in a knowledge base, and in turn uses the knowledge base to interpret and link newly extracted information with respect to already existing entities.
This paper investigates the relation between TT-MCTAG, a formalism used in computational linguistics, and RCG. RCGs are known to describe exactly the class PTIME; simple RCG even have been shown to be equivalent to linear context-free rewriting systems, i.e., to be mildly context-sensitive. TT-MCTAG has been proposed to model free word order languages. In general, it is NP-complete. In this paper, we will put an additional limitation on the derivations licensed in TT-MCTAG. We show that TT-MCTAG with this additional limitation can be transformed into equivalent simple RCGs. This result is interesting for theoretical reasons (since it shows that TT-MCTAG in this limited form is mildly context-sensitive) and, furthermore, even for practical reasons: We use the proposed transformation from TT-MCTAG to RCG in an actual parser that we have implemented.
The author presents MASSY, the MODULAR AUDIOVISUAL SPEECH SYNTHESIZER. The system combines two approaches of visual speech synthesis. Two control models are implemented: a (data based) di-viseme model and a (rule based) dominance model where both produce control commands in a parameterized articulation space. Analogously two visualization methods are implemented: an image based (video-realistic) face model and a 3D synthetic head. Both face models can be driven by both the data based and the rule based articulation model.
The high-level visual speech synthesis generates a sequence of control commands for the visible articulation. For every virtual articulator (articulation parameter) the 3D synthetic face model defines a set of displacement vectors for the vertices of the 3D objects of the head. The vertices of the 3D synthetic head then are moved by linear combinations of these displacement vectors to visualize articulation movements. For the image based video synthesis a single reference image is deformed to fit the facial properties derived from the control commands. Facial feature points and facial displacements have to be defined for the reference image. The algorithm can also use an image database with appropriately annotated facial properties. An example database was built automatically from video recordings. Both the 3D synthetic face and the image based face generate visual speech that is capable to increase the intelligibility of audible speech.
Other well known image based audiovisual speech synthesis systems like MIKETALK and VIDEO REWRITE concatenate pre-recorded single images or video sequences, respectively. Parametric talking heads like BALDI control a parametric face with a parametric articulation model. The presented system demonstrates the compatibility of parametric and data based visual speech synthesis approaches.
The problem of vocalization, or diacritization, is essential to many tasks in Arabic NLP. Arabic is generally written without the short vowels, which leads to one written form having several pronunciations with each pronunciation carrying its own meaning(s). In the experiments reported here, we define vocalization as a classification problem in which we decide for each character in the unvocalized word whether it is followed by a short vowel. We investigate the importance of different types of context. Our results show that the combination of using memory-based learning with only a word internal context leads to a word error rate of 6.64%. If a lexical context is added, the results deteriorate slowly.
This paper sets up a framework for LTAG (Lexicalized Tree Adjoining Grammar) semantics that brings together ideas from different recent approaches addressing some shortcomings of TAG semantics based on the derivation tree. Within this framework, several sample analyses are proposed, and it is shown that the framework allows to analyze data that have been claimed to be problematic for derivation tree based LTAG semantics approaches.
LTAG semantics for questions
(2004)
This papers presents a compositional semantic analysis of interrogatives clauses in LTAG (Lexicalized Tree Adjoining Grammar) that captures the scopal properties of wh- and nonwh-quantificational elements. It is shown that the present approach derives the correct semantics for examples claimed to be problematic for LTAG semantic approaches based on the derivation tree. The paper further provides an LTAG semantics for embedded interrogatives.
A lot of interest has recently been paid to constraint-based definitions and extensions of Tree Adjoining Grammars (TAG). Examples are the so-called quasi-trees, D-Tree Grammars and Tree Description Grammars. The latter are grammars consisting of a set of formulars denoting trees. TDGs are derivation based where in each derivation step a conjunction is built of the old formular, a formular of the grammar and additional equivalences between node names of the two formulars. This formalism is more powerfull than TAGs. TDGs offer the advantages of MC-TAG and D-Tree Grammars for natural languages and they allow underspecification. However the problem is that TDGs might be unnecessarily powerfull for natural languages. To solve this problem, in this paper, I will propose a local TDGs, a restricted version of TDGs. Local TDGs still have the advantages of TDGs but they are semilinear and therefore more appropriate for natural languages. First, the notion of the semilinearity is defined. Then local TDGs are introduced, and, finally, semilinearity of local Tree Description Languages is proven.
The goal of our current project is to build a system that can learn to imitate a version of a spoken utterance using an articulatory speech synthesiser. The approach is informed and inspired by knowledge of early infant speech development. Thus we expect our system to reproduce and exploit the utility of infant behaviours such as listening, vocal play, babbling and word imitation. We expect our system to develop a relationship between the sound-making capabilities of its vocal tract and the phonetic/phonological structure of imitated utterances. At the heart of our approach is the learning of an inverse model that relates acoustic and motor representations of speech. The acoustic to auditory mappings uses an auditory filter bank and a self-organizing phase of learning. The inverse model from auditory to vocal tract control parameters is estimated using a babbling phase, in which the vocal tract is essentially driven in a random manner, much like the babbling phase of speech acquisition in infants. The complete system can be used to imitate simple utterances through a direct mapping from sound to control parameters. Our initial results show that this procedure works well for sounds generated by its own voice. Further work is needed to build a phonological control level and achieve better performance with real speech.
In syntax, the trend nowadays is towards lexicalized grammar formalisms. It is now widely accepted that dividing words into wordclasses may serve as a laborsaving mechanism - but at the same time, it discards all detailed information on the idiosyncratic behavior of words. And that is exactly the type of information that may be necessary in order to parse a sentence. For learning approaches, however, lexicalized grammars represent a challenge for the very reason that they include so much detailed and specific information, which is difficult to learn. This paper will present an algorithm for learning a link grammar of German. The problem of data sparseness is tackled by using all the available information from partial parses as well as from an existing grammar fragment and a tagger. This is a report about work in progress so there are no representative results available yet.
A comprehensive investigation of Japanese particle was missing up to now. General implications were set up without the fact that a comprehensive analysis was carried out. [...] We offer a lexicalist treatment of the problem. Instead of assuming different phrase structure rules we state a type hierarchy of Japanese particles. This makes a uniform treatment of phrase structure as well as a differentiation of subcategorization patterns possible.
We present a solution for the representation of Japanese honorifical information in the HPSG framework. Basically, there are three dimensions of honorification. We show that a treatment is necessary that involves both the syntactic and the contextual level of information. The japanese grammar is part of a machine translation system.
In this text, we describe the development of a broad coverage grammar for Japanese that has been built for and used in different application contexts. The grammar is based on work done in the Verbmobil project (Siegel 2000) on machine translation of spoken dialogues in the domain of travel planning. The second application for JACY was the automatic email response task. Grammar development was described in Oepen et al. (2002a). Third, it was applied to the task of understanding material on mobile phones available on the internet, while embedded in the project DeepThought (Callmeier et al. 2004, Uszkoreit et al. 2004). Currently, it is being used for treebanking and ontology extraction from dictionary definition sentences by the Japanese company NTT (Bond et al. 2004).
This paper presents a comparative study of probabilistic treebank parsing of German, using the Negra and TüBa-D/Z treebanks. Experiments with the Stanford parser, which uses a factored PCFG and dependency model, show that, contrary to previous claims for other parsers, lexicalization of PCFG models boosts parsing performance for both treebanks. The experiments also show that there is a big difference in parsing performance, when trained on the Negra and on the TüBa-D/Z treebanks. Parser performance for the models trained on TüBa-D/Z are comparable to parsing results for English with the Stanford parser, when trained on the Penn treebank. This comparison at least suggests that German is not harder to parse than its West-Germanic neighbor language English.
While the sortal constraints associated with Japanese numeral classifiers are wellstudied, less attention has been paid to the details of their syntax. We describe an analysis implemented within a broadcoverage HPSG that handles an intricate set of numeral classifier construction types and compositionally relates each to an appropriate semantic representation, using Minimal Recursion Semantics.
While the sortal constraints associated with Japanese numeral classifiers are well-studied, less attention has been paid to the details of their syntax. We describe an analysis implemented within a broad-coverage HPSG that handles an intricate set of numeral classifier construction types and compositionally relates each to an appropriate semantic representation, using Minimal Recursion Semantics.
Hybrid robust deep and shallow semantic processing for creativity support in document production
(2004)
The research performed in the DeepThought project (http://www.project-deepthought.net) aims at demonstrating the potential of deep linguistic processing if added to existing shallow methods that ensure robustness. Classical information retrieval is extended by high precision concept indexing and relation detection. We use this approach to demonstrate the feasibility of three ambitious applications, one of which is a tool for creativity support in document production and collective brainstorming. This application is described in detail in this paper. Common to all three applications, and the basis for their development is a platform for integrated linguistic processing. This platform is based on a generic software architecture that combines multiple NLP components and on robust minimal recursive semantics (RMRS) as a uniform representation language.
How to compare treebanks
(2008)
Recent years have seen an increasing interest in developing standards for linguistic annotation, with a focus on the interoperability of the resources. This effort, however, requires a profound knowledge of the advantages and disadvantages of linguistic annotation schemes in order to avoid importing the flaws and weaknesses of existing encoding schemes into the new standards. This paper addresses the question how to compare syntactically annotated corpora and gain insights into the usefulness of specific design decisions. We present an exhaustive evaluation of two German treebanks with crucially different encoding schemes. We evaluate three different parsers trained on the two treebanks and compare results using EVALB, the Leaf-Ancestor metric, and a dependency-based evaluation. Furthermore, we present TePaCoC, a new testsuite for the evaluation of parsers on complex German grammatical constructions. The testsuite provides a well thought-out error classification, which enables us to compare parser output for parsers trained on treebanks with different encoding schemes and provides interesting insights into the impact of treebank annotation schemes on specific constructions like PP attachment or non-constituent coordination.
In the last decade, the Penn treebank has become the standard data set for evaluating parsers. The fact that most parsers are solely evaluated on this specific data set leaves the question unanswered how much these results depend on the annotation scheme of the treebank. In this paper, we will investigate the influence which different decisions in the annotation schemes of treebanks have on parsing. The investigation uses the comparison of similar treebanks of German, NEGRA and TüBa-D/Z, which are subsequently modified to allow a comparison of the differences. The results show that deleted unary nodes and a flat phrase structure have a negative influence on parsing quality while a flat clause structure has a positive influence.
Japanese is often taken to be strictly head-final in its syntax. In our work on a broad-coverage, precision implemented HPSG for Japanese, we have found that while this is generally true, there are nonetheless a few minor exceptions to the broad trend. In this paper, we describe the grammar engineering project, present the exceptions we have found, and conclude that this kind of phenomenon motivates on the one hand the HPSG type hierarchical approach which allows for the statement of both broad generalizations and exceptions to those generalizations and on the other hand the usefulness of grammar engineering as a means of testing linguistic hypotheses.
This demo abstract describes the SmartWeb Ontology-based Information Extraction System (SOBIE). A key feature of SOBIE is that all information is extracted and stored with respect to the SmartWeb ontology. In this way, other components of the systems, which use the same ontology, can access this information in a straightforward way. We will show how information extracted by SOBIE is visualized within its original context, thus enhancing the browsing experience of the end user.
In the past, a divide could be seen between ’deep’ parsers on the one hand, which construct a semantic representation out of their input, but usually have significant coverage problems, and more robust parsers on the other hand, which are usually based on a (statistical) model derived from a treebank and have larger coverage, but leave the problem of semantic interpretation to the user. More recently, approaches have emerged that combine the robustness of datadriven (statistical) models with more detailed linguistic interpretation such that the output could be used for deeper semantic analysis. Cahill et al. (2002) use a PCFG-based parsing model in combination with a set of principles and heuristics to derive functional (f-)structures of Lexical-Functional Grammar (LFG). They show that the derived functional structures have a better quality than those generated by a parser based on a state-of-the-art hand-crafted LFG grammar. Advocates of Dependency Grammar usually point out that dependencies already are a semantically meaningful representation (cf. Menzel, 2003). However, parsers based on dependency grammar normally create underspecified representations with respect to certain phenomena such as coordination, apposition and control structures. In these areas they are too "shallow" to be directly used for semantic interpretation. In this paper, we adopt a similar approach to Cahill et al. (2002) using a dependency-based analysis to derive functional structure, and demonstrate the feasibility of this approach using German data. A major focus of our discussion is on the treatment of coordination and other potentially underspecified structures of the dependency data input. F-structure is one of the two core levels of syntactic representation in LFG (Bresnan, 2001). Independently of surface order, it encodes abstract syntactic functions that constitute predicate argument structure and other dependency relations such as subject, predicate, adjunct, but also further semantic information such as the semantic type of an adjunct (e.g. directional). Normally f-structure is captured as a recursive attribute value matrix, which is isomorphic to a directed graph representation. Figure 5 depicts an example target f-structure. As mentioned earlier, these deeper-level dependency relations can be used to construct logical forms as in the approaches of van Genabith and Crouch (1996), who construct underspecified discourse representations (UDRSs), and Spreyer and Frank (2005), who have robust minimal recursion semantics (RMRS) as their target representation. We therefore think that f-structures are a suitable target representation for automatic syntactic analysis in a larger pipeline of mapping text to interpretation. In this paper, we report on the conversion from dependency structures to fstructure. Firstly, we evaluate the f-structure conversion in isolation, starting from hand-corrected dependencies based on the TüBa-D/Z treebank and Versley (2005)´s conversion. Secondly, we start from tokenized text to evaluate the combined process of automatic parsing (using Foth and Menzel (2006)´s parser) and f-structure conversion. As a test set, we randomly selected 100 sentences from TüBa-D/Z which we annotated using a scheme very close to that of the TiGer Dependency Bank (Forst et al., 2004). In the next section, we sketch dependency analysis, the underlying theory of our input representations, and introduce four different representations of coordination. We also describe Weighted Constraint Dependency Grammar (WCDG), the dependency parsing formalism that we use in our experiments. Section 3 characterises the conversion of dependencies to f-structures. Our evaluation is presented in section 4, and finally, section 5 summarises our results and gives an overview of problems remaining to be solved.
Transforming constituent-based annotation into dependency-based annotation has been shown to work for different treebanks and annotation schemes (e.g. Lin (1995) has transformed the Penn treebank, and Kübler and Telljohann (2002) the Tübinger Baumbank des Deutschen (TüBa-D/Z)). These ventures are usually triggered by the conflict between theory-neutral annotation, that targets most needs of a wider audience, and theory-specific annotation, that provides more fine-grained information for a smaller audience. As a compromise, it has been pointed out that treebanks can be designed to support more than one theory from the start (Nivre, 2003). We argue that information can also be added to an existing annotation scheme so that it supports additional theory-specific annotations. We also argue that such a transformation is useful for improving and extending the original annotation scheme with respect to both ambiguous annotation and annotation errors. We show this by analysing problems that arise when generating dependency information from the constituent-based TüBa-D/Z.
Chunk parsing has focused on the recognition of partial constituent structures at the level of individual chunks. Little attention has been paid to the question of how such partial analyses can be combined into larger structures for complete utterances. Such larger structures are not only desirable for a deeper syntactic analysis. They also constitute a necessary prerequisite for assigning function-argument structure. The present paper offers a similaritybased algorithm for assigning functional labels such as subject, object, head, complement, etc. to complete syntactic structures on the basis of prechunked input. The evaluation of the algorithm has concentrated on measuring the quality of functional labels. It was performed on a German and an English treebank using two different annotation schemes at the level of function argument structure. The results of 89.73% correct functional labels for German and 90.40%for English validate the general approach.
The process of turning a hand-written HPSG theory into a working computational grammar requires complex considerations. Two leading platforms are available for implementing HPSG grammars: The LKB and TRALE. These platforms are based on different approaches, distinct in their underlying logics and implementation details. This paper adopts the perspective of a computational linguist whose goal is to implement an HPSG theory. It focuses on ten different dimensions, relevant to HPSG grammar implementation, and examines, compares, and evaluates the different means which the two approaches provide for implementing them. The paper concludes that the approaches occupy opposite positions on two axes: faithfulness to the hand-written theory and computational accessibility. The choice between them depends largely on the grammar writer's preferences regarding those properties.
This paper addresses the problem ofconstraints for relative quantifier sope, in partiular in inverse linking readings wherecertain scope orders are exluded. We show how to account for such restrictions in the Tree Adjoining Grammar (TAG) framework by adopting a notion offlexible composition. In the semantics we use for TAG we introduce quantifier sets that group quantifiers that are "glued" together in the sense that no other quantifieran scopally intervene between them. Theflexible composition approach allows us to obtain the desired quantifier sets and thereby the desiredconstraints for quantifier sope.
In this paper we will explore the similarities and differences between two feature logic-based approaches to the composition of semantic representations. The first approach is formulated for Lexicalized Tree Adjoining Grammar (LTAG, Joshi and Schabes 1997), the second is Lexical Ressource Semantics (LRS, Richter and Sailer 2004) and was first defined in Head-driven Phrase Structure Grammar. The two frameworks have several common characteristics that make them easy to compare: 1 They use languages of two sorted type theory for semantic representations. 2. They allow underspecification. LTAG uses scope constraints while LRS provides component-of contraints. 3 They use feature logics for computing semantic representations. 4. they are designed for computational applications. By comparing the two frameworks we will also point outsome characteristics and advantages of feature logic-based semantic computation in genereal.
TT-MCTAG lets one abstract away from the relative order of co-complements in the final derived tree, which is more appropriate than classic TAG when dealing with flexible word order in German. In this paper, we present the analyses for sentential complements, i.e., wh-extraction, thatcomplementation and bridging, and we work out the crucial differences between these and respective accounts in XTAG (for English) and V-TAG (for German).
This paper proposes a compositional semantics for lexicalized tree adjoining grammars (LTAG). Tree-local multicompnent derivations allow seperation of semantiv contribution of a lexical item into one component contributing to the predicate argument structure and second a component contributing to scope semantics. Based on this idea a syntx-semantics interface is presented where the compositional semantics depends only on the derivation structure. It is shown that the derivation structure allows an appropriate amount of underspecification. This is illustrated by investigating underspecified representations for quantifier scpoe ambiguities and related phenomena such as adjunct scope and island constraints.
In this paper we propose a compositional semantics for lexicalized tree-adjoining grammar (LTAG). Tree-local multicomponent derivations allow separation of the semantic contribution of a lexical item into one component contributing to the predicate argument structure and a second component contributing to scope semantics. Based on this idea a syntax-semantics interface is presented where the compositional semantics depends only on the derivation structure. It is shown that the derivation structure (and indirectly the locality of derivations) allows an appropriate amount of underspecification. This is illustrated by investigating underspecified representations for quantifier scope ambiguities and related phenomena such as adjunct scope and island constraints.
This paper is concerned with the tagging of spatial expressions in German newspaper articles, assigning a meaning to the expression and classifying the usages of the spatial expression and linking the derived referent to an event description. In our system, we implemented the activation of concepts in a very simple fashion, a concept is activated once (with a cost depending on the item that activated it) and is left activated thereafter. As an example, a city also activates the nodes for the region and the country it is part of, so that cities from one country are chosen over cities from different countries. A test corpus of 12 German newspaper articles was tested regarding several disambiguation strategies. Disambiguation was carried out via a beam search to find an approximately cost-optimal solution for the conflict set of potential grounding candidates for the tagged spatial expression. Test showed that the disambiguation strategies improved accuracy significantly.
In this paper, we investigate the role of sub-optimality in training data for part-of-speech tagging. In particular, we examine to what extent the size of the training corpus and certain types of errors in it affect the performance of the tagger. We distinguish four types of errors: If a word is assigned a wrong tag, this tag can belong to the ambiguity class of the word (i.e. to the set of possible tags for that word) or not; furthermore, the major syntactic category (e.g. "N" or "V") can be correctly assigned (e.g. if a finite verb is classified as an infinitive) or not (e.g. if a verb is classified as a noun). We empirically explore the decrease of performance that each of these error types causes for different sizes of the training set. Our results show that those types of errors that are easier to eliminate have a particularly negative effect on the performance. Thus, it is worthwhile concentrating on the elimination of these types of errors, especially if the training corpus is large.
We present a broad coverage Japanese grammar written in the HPSG formalism with MRS semantics. The grammar is created for use in real world applications, such that robustness and performance issues play an important role. It is connected to a POS tagging and word segmentation tool. This grammar is being developed in a multilingual context, requiring MRS structures that are easily comparable across languages.
Using a qualitative analysis of disagreements from a referentially annotated newspaper corpus, we show that, in coreference annotation, vague referents are prone to greater disagreement. We show how potentially problematic cases can be dealt with in a way that is practical even for larger-scale annotation, considering a real-world example from newspaper text.
Dialogue acts in Verbmobil 2
(1998)
This report describes the dialogue phases and the second edition dialogue acts which are used in the VERBMOBIL 2 project [...]. While in the first project phase the scenario was restricted to appointment scheduling dialogues, it has been extended to travel planning in the second phase with appointment scheduling being only a part of the new scenario.
Developing linguistic resources, in particular grammars, is known to be a complex task in itself, because of (amongst others) redundancy and consistency issues. Furthermore some languages can reveal themselves hard to describe because of specific characteristics, e.g. the free word order in German. In this context, we present (i) a framework allowing to describe tree-based grammars, and (ii) an actual fragment of a core multicomponent tree-adjoining grammar with tree tuples (TT-MCTAG) for German developed using this framework. This framework combines a metagrammar compiler and a parser based on range concatenation grammar (RCG) to respectively check the consistency and the correction of the grammar. The German grammar being developed within this framework already deals with a wide range of scrambling and extraction phenomena.
Das Problem des Transfers in der maschinellen Übersetzung von Japanisch nach Englisch ist fehlende Information über Numerus und Definitheit im Japanischen, die für die Wahl der englischen Artikel und die Nomenmarkierung gebraucht wird. Obwohl dieses Problem signifikant ist, beschäftigt sich die Forschungsliteratur kaum damit. [...] Wir bsaieren unsere Untersuchungen auf experimentell erhobenen Daten aus einem Experiment über deutsch-japanische gedolmetschte Terminaushandlungsdialoge [...]. Auf diese Weise können Phänomene bestimmt werden, die für die Domäne von VERBMOBIL relevant sind. Wir sehen unser Vorgehen in Übereinstimmung mit dem 'Sublanguage'-Ansatz [...].
Eins der signifikanten Probleme in der maschinellen Übersetzung japanische in deutsche Sprache ist die fehlende Information und Definitheit im japanischen Analyse-Output. Eine effiziente Lösung dieses Problems ist es, die Suche nach der relevanten Information in den Transfer zu integrieren. Transferregeln werden mit Präferenzregeln und Default-Regeln kombiniert. Dadurch wird Information über lexikalische Restriktionen der Zielsprache, über die Domäne und über den Diskurs zugänglich.
In this paper we show an approach to the customization of GermaNet to the German HPSG grammar lexicon developed in the Verbmobil project. GermaNet has a broad coverage of the German base vocabulary and fine-grained semantic classification; while the HPSG grammar lexicon is comparatively small und has a coarse-grained semantic classification. In our approach, we have developed a mapping algorithm to relate the synsets in GermaNet with the semantic sorts in HPSG. The evaluation result shows that this approach is useful for the lexical extension of our deep grammar development to cope with real-world text understanding.
We present an effort for the development of multilingual named entity grammars in a unification-based finite-state formalism (SProUT). Following an extended version of the MUC7 standard, we have developed Named Entity Recognition grammars for German, Chinese, Japanese, French, Spanish, English, and Czech. The grammars recognize person names, organizations, geographical locations, currency, time and date expressions. Subgrammars and gazetteers are shared as much as possible for the grammars of the different languages. Multilingual corpora from the business domain are used for grammar development and evaluation. The annotation format (named entity and other linguistic information) is described. We present an evaluation tool which provides detailed statistics and diagnostics, allows for partial matching of annotations, and supports user-defined mappings between different annotation and grammar output formats.
This paper compares two approaches to computational semantics, namely semantic unification in Lexicalized Tree Adjoining Grammars (LTAG) and Lexical Resource Semantics (LRS) in HPSG. There are striking similarities between the frameworks that make them comparable in many respects. We will exemplify the differences and similarities by looking at several phenomena. We will show, first of all, that many intuitions about the mechanisms of semantic computations can be implemented in similar ways in both frameworks. Secondly, we will identify some aspects in which the frameworks intrinsically differ due to more general differences between the approaches to formal grammar adopted by LTAG and HPSG.
The two papers included in this volume have developed from work with the CHILDES tools and the Media Editor in the two research projects, "Second language acquisition of German by Russian learners", sponsored by the Max Planck Institute for Psycholinguistics, Nijmegen, from 1998 to 1999 (directed by Ursula Stephany, University of Cologne, and Wolfgang Klein, Max Planck Institute for Psycholinguistics, Nijmegen) and "The age factor in the acquisition of German as a second language", sponsored by the German Science Foundation (DFG), Bonn, since 2000 (directed by Ursula Stephany, University of Cologne, and Christine Dimroth, Max Planck Institute for Psycholinguistics, Nijmegen). The CHILDES Project has been developed and is being continuously improved at Carnegie Mellon University, Pittsburgh, under the supervision of Brian MacWhinney. Having used the CHILDES tools for more than ten years for transcribing and analyzing Greek child data there it was no question that I would also use them for research into the acquisition of German as a second language and analyze the big amount of spontaneous speech gathered from two Russian girls with the help of the CLAN programs. When in the spring of 1997, Steven Gillis from the University of Antwerp (in collaboration with Gert Durieux) developed a lexicon-based automatic coding system based on the CLAN program MOR and suitable for coding languages with richer morphologies than English, such as Modern Greek. Coding huge amounts of data then became much quicker and more comfortable so that I decided to adopt this system for German as well. The paper "Working with the CHILDES Tools" is based on two earlier manuscripts which have grown out of my research on Greek child language and the many CHILDES workshops taught in Germany, Greece, Portugal, and Brazil over the years. Its contents have now been adapted to the requirements of research into the acquisition of German as a second language and for use on Windows.
Prepositional phrase (PP) attachment is one of the major sources for errors in traditional statistical parsers. The reason for that lies in the type of information necessary for resolving structural ambiguities. For parsing, it is assumed that distributional information of parts-of-speech and phrases is sufficient for disambiguation. For PP attachment, in contrast, lexical information is needed. The problem of PP attachment has sparked much interest ever since Hindle and Rooth (1993) formulated the problem in a way that can be easily handled by machine learning approaches: In their approach, PP attachment is reduced to the decision between noun and verb attachment; and the relevant information is reduced to the two possible attachment sites (the noun and the verb) and the preposition of the PP. Brill and Resnik (1994) extended the feature set to the now standard 4-tupel also containing the noun inside the PP. Among many publications on the problem of PP attachment, Volk (2001; 2002) describes the only system for German. He uses a combination of supervised and unsupervised methods. The supervised method is based on the back-off model by Collins and Brooks (1995), the unsupervised part consists of heuristics such as ”If there is a support verb construction present, choose verb attachment”. Volk trains his back-off model on the Negra treebank (Skut et al., 1998) and extracts frequencies for the heuristics from the ”Computerzeitung”. The latter also serves as test data set. Consequently, it is difficult to compare Volk’s results to other results for German, including the results presented here, since not only he uses a combination of supervised and unsupervised learning, but he also performs domain adaptation. Most of the researchers working on PP attachment seem to be satisfied with a PP attachment system; we have found hardly any work on integrating the results of such approaches into actual parsers. The only exceptions are Mehl et al. (1998) and Foth and Menzel (2006), both working with German data. Mehl et al. report a slight improvement of PP attachment from 475 correct PPs out of 681 PPs for the original parser to 481 PPs. Foth and Menzel report an improvement of overall accuracy from 90.7% to 92.2%. Both integrate statistical attachment preferences into a parser. First, we will investigate whether dependency parsing, which generally uses lexical information, shows the same performance on PP attachment as an independent PP attachment classifier does. Then we will investigate an approach that allows the integration of PP attachment information into the output of a parser without having to modify the parser: The results of an independent PP attachment classifier are integrated into the parse of a dependency parser for German in a postprocessing step.
We investigate methods to improve the recall in coreference resolution by also trying to resolve those definite descriptions where no earlier mention of the referent shares the same lexical head (coreferent bridging). The problem, which is notably harder than identifying coreference relations among mentions which have the same lexical head, has been tackled with several rather different approaches, and we attempt to provide a meaningful classification along with a quantitative comparison. Based on the different merits of the methods, we discuss possibilities to improve them and show how they can be effectively combined.
This report explores the question of compatibility between annotation projects including translating annotation formalisms to each other or to common forms. Compatibility issues are crucial for systems that use the results of multiple annotation projects. We hope that this report will begin a concerted effort in the field to track the compatibility of annotation schemes for part of speech tagging, time annotation, treebanking, role labeling and other phenomena.