Refine
H-BRS Bibliography
- yes (159) (remove)
Departments, institutes and facilities
- Fachbereich Informatik (46)
- Präsidium (35)
- Fachbereich Wirtschaftswissenschaften (25)
- Institute of Visual Computing (IVC) (18)
- Fachbereich Angewandte Naturwissenschaften (17)
- Fachbereich Ingenieurwissenschaften und Kommunikation (17)
- Institut für funktionale Gen-Analytik (IFGA) (12)
- Fachbereich Sozialpolitik und Soziale Sicherung (10)
- Institut für Cyber Security & Privacy (ICSP) (2)
- Institut für Sicherheitsforschung (ISF) (2)
Document Type
- Conference Object (41)
- Article (36)
- Part of Periodical (33)
- Part of a Book (16)
- Book (monograph, edited volume) (11)
- Report (8)
- Master's Thesis (5)
- Bachelor Thesis (3)
- Conference Proceedings (3)
- Contribution to a Periodical (2)
- Working Paper (1)
Year of publication
- 2011 (159) (remove)
Keywords
- Unternehmen (3)
- CUDA (2)
- Emergency support system (2)
- Finite-Elemente-Methode (2)
- Mobile sensors (2)
- Robotik (2)
- 3D Crosstalk (1)
- 3D Display (1)
- 3D Visualisierung (1)
- 3D crosstalk (1)
Ordnung des Institutes für Detektionstechnologie an der Hochschule Bonn-Rhein-Sieg vom 23.09.2010
(2011)
Wohin in der Reha?
(2011)
Arbeitsmarktintegration - eine Aufgabe der medizinischen Rehabilitation Abhängigkeitskranker?
(2011)
The recent explosion of available audio-visual media is the new challenge for information retrieval research. Audio speech recognition systems translate spoken content to the text domain. There is a need for searching and indexing this data which possesses no logical structure. One possible way to structure it on a high level of abstraction is by finding topic boundaries. Two unsupervised topic segmentation methods were evaluated with real-world data in the course of this work. The first one, TSF, models topic shifts as fluctuations in the similarity function of the transcript. The second one, LCSeg, approaches topic changes as places with the least overlapping lexical chains. Only LCSeg performed close to a similar real-world corpus. Other reported results could not be outperformed. Topic analysis based on the repeated word usage models renders topic changes more ambiguous than expected. This issue has more impact on the segmentation quality than the state-of-the-art ASR word error rate. It could be concluded that it is advisable to develop topic segmentation algorithms with real-world data to avoid potential biases to artificial data. Unlike evaluated approaches based on word usage analysis, methods operating with local contexts can be expected to perform better through emulation of semantic dependencies.