Refine
Document Type
- Conference Proceeding (3) (remove)
Has Fulltext
- yes (3)
Is part of the Bibliography
- no (3)
Keywords
Publicationstate
- Veröffentlichungsversion (3) (remove)
Reviewstate
Publisher
Sowohl bei der Entwicklung konventioneller Software als auch bei der Entwicklung wissensbasierter Systeme fehlen z.Z. systematische Ansätze, Anforderungen an das zu entwickelnde Produkt „ingenieurmäßig“ zu erheben. Die Probleme, mit denen sich der Software Engineer konfrontiert sieht, ähneln denen der Wissensakquisition im Knowledge Engineering. Der an der Universität Karlsruhe am Institut AIFB entwickelte Ansatz MIKE ([AFL93]) beschreibt eine systematische Vorgehensweise zur Entwicklung wissensbasierter Systeme. Die Beschreibung der spezifischen Anforderungen an wissensbasierte Systeme ist Gegenstand der aktuellen Forschung; mit MIKE steht aber bereits das Gerüst zur Verfügung, mit denen die Anforderungen im Laufe der weiteren Entwicklungsphasen verwaltet werden können.
Empirical synchronic language studies generally seek to investigate language phenomena for one point in time, even though this point in time is often not stated explicitly. Until today, surprisingly little research has addressed the implications of this time-dependency of synchronic research on the composition and analysis of data that are suitable for conducting such studies. Existing solutions and practices tend to be too general to meet the needs of all kinds of research questions. In this theoretical paper that is targeted at both corpus creators and corpus users, we propose to take a decidedly synchronic perspective on the relevant language data. Such a perspective may be realised either in terms of sampling criteria or in terms of analytical methods applied to the data. As a general approach for both realisations, we introduce and explore the FReD strategy (Frequency Relevance Decay) which models the relevance of language events from a synchronic perspective. This general strategy represents a whole family of synchronic perspectives that may be customised to meet the requirements imposed by the specific research questions and language domain under investigation.
The central issue in corpus-driven linguistics is the detection and description of patterns in language usage. The features that constitute the notion of a pattern can be computed to a certain extent by statistical (collocation) methods, but a crucial part of the notion may vary depending on applications and users. Thus, typically, any computed collocation cluster will have to be interpreted hermeneutically. Often it might be captured by a generalized, more abstract pattern. We present a generic process model that supports the recognition, interpretation, and expression of the patterns inside and of the relations between clusters. By this, clusters can be merged virtually according to any notion of a 'pattern', and their relations can be exploited for different applications