TU Berlin

Fachgebiet Datenbanksysteme und InformationsmanagementPublikationen

Logo FG DIMA-new  65px


zur Navigation


SemAcom: A System for Modeling with Semantic Autocompletion
Zitatschl├╝ssel henning-agt-bizware-models-2012-2
Autor Henning Agt
Buchtitel Model Driven Engineering Languages and Systems - Demo Track, 15th International Conference, MODELS 2012, Innsbruck, Austria, Sept. 30 - Oct. 05, 2012
Jahr 2012
Zusammenfassung Autocompletion is a feature in various applications that assists users while typing into input fields. It predicts the user input based on background knowledge, rules and heuristics and is successfully used in mobile phones, web browsers, search engines and integrated development environments. We demonstrate SemAcom, a system for automatically suggesting related terms for domain-specific models. During the development of Ecore models with the Eclipse Modeling Framework SemAcom provides context-sensitive pop-up lists with related class names. The backbone of our system is a large-scale network of semantically related terms that was created automatically from a natural language dataset. The network contains probabilistic information how strong terms are related and is used to retrieve ranked lists of semantically related terms for a given set of input terms. That allows us to present the most relevant terms at the top of the list of suggestions.
Link zur Publikation Link zur Originalpublikation Download Bibtex Eintrag



Schnellnavigation zur Seite über Nummerneingabe