Refine
Year of publication
- 2013 (5) (remove)
Document Type
- Part of a Book (3)
- Article (2)
Has Fulltext
- yes (5)
Is part of the Bibliography
- no (5)
Keywords
- Computer-Mediated Communication (2)
- Corpus Linguistics (2)
- Decision Trees (2)
- Genitive Classification (2)
- Grammar (2)
- Machine Leaming (2)
- Arzt (1)
- Bootstrapping methods (1)
- Brustkrebs (1)
- Case (1)
Publicationstate
- Postprint (2)
- Veröffentlichungsversion (1)
Reviewstate
- (Verlags)-Lektorat (2)
- Peer-Review (1)
Publisher
- Springer (5) (remove)
The present paper provides a new approach to the form-function relation in Latin declension. First, inflections are discussed from a functional point of view with special consideration to questions of syncretism. A case hierarchy is justified for Latin that conforms to general observations on case systems. The analysis leads to a markedness scale that provides a ranking of case-number-combinations from unmarked to most marked. Systematic syncretism always applies to contiguous sections of the case-number-scale (‘syncretism fields’). Second, inflections are analysed from a formal point of view taking into account partial identities and differences among noun endings. Theme vowels being factored out, endings are classified on the basis of their make-up, e.g., as sigmatic endings; as containing desinential (non-thematic) vowels; as containing long vowels; and so on. The analysis leads to a view of endings as involving more basic elements or ‘markers’. Endings of the various declensions instantiate a small number of types, and these can be put into a ranked order (a formal scale) that applies transparadigmatically. Third, the relationship between the independently substantiated functional and formal hierarchies is examined. In any declension, the form-function-relationship is established by aligning the relevant formal and functional scales (or ‘sequences’). Some types of endings are in one-to-one correspondence with bundles of morphosyntactic properties as they should be according to a classical morphemic approach, but others are not. Nevertheless, endings can be assigned a uniform role if the form-function-relationship is understood to be based on an alignment of formal and functional sequences. A diagrammatical form-function relationship is revealed that could not be captured in classical or refined morphemic approaches.
Contemporary studies on the characteristics of natural language benefit enormously from the increasing amount of linguistic corpora. Aside from text and speech corpora, corpora of computer-mediated communication (CMC) position themselves between orality and literacy, and beyond that provide insight into the impact of “new”, mainly internet-based media on language behaviour. In this paper, we present an empirical attempt to work with annotated CMC corpora for the explanation of linguistic phenomena. In concrete terms, we implement machine learning algorithms to produce decision trees that reveal rules and tendencies about the use of genitive markers in German.
Contemporary studies on the characteristics of natural language benefit enormously from the increasing amount of linguistic corpora. Aside from text and speech corpora, corpora of computer-mediated communication (CMC) Position themselves between orality and literacy, and beyond that provide in- sight into the impact of "new", mainly intemet-based media on language beha- viour. In this paper, we present an empirical attempt to work with annotated CMC corpora for the explanation of linguistic phenomena. In concrete terms, we implement machine leaming algorithms to produce decision trees that reveal rules and tendencies about the use of genitive markers in German.
In this article, we examine the effectiveness of bootstrapping supervised machine-learning polarity classifiers with the help of a domain-independent rule-based classifier that relies on a lexical resource, i.e., a polarity lexicon and a set of linguistic rules. The benefit of this method is that though no labeled training data are required, it allows a classifier to capture in-domain knowledge by training a supervised classifier with in-domain features, such as bag of words, on instances labeled by a rule-based classifier. Thus, this approach can be considered as a simple and effective method for domain adaptation. Among the list of components of this approach, we investigate how important the quality of the rule-based classifier is and what features are useful for the supervised classifier. In particular, the former addresses the issue in how far linguistic modeling is relevant for this task. We not only examine how this method performs under more difficult settings in which classes are not balanced and mixed reviews are included in the data set but also compare how this linguistically-driven method relates to state-of-the-art statistical domain adaptation.