Article
Refine
Year of publication
- 2024 (16) (remove)
Document Type
- Article (16) (remove)
Keywords
- Hot S-parameter (2)
- 1P hub loads (1)
- 3-D printing (1)
- Acyl-amino acids (1)
- Acylation (1)
- Aeroelasticity (1)
- Ageing (1)
- Aminoacylase (1)
- Anatomy (1)
- Biocatalysis (1)
- Biosurfactants (1)
- Capacitive model (1)
- Chaperone (1)
- Circuit simulation (1)
- Complex-valued eigenvalues (1)
- Constitutive model (1)
- Damage mechanics theory (1)
- Discontinuous fractures (1)
- Eigenvalue trajectories (1)
- Enzyme coverage (1)
Institute
- Fachbereich Medizintechnik und Technomathematik (9)
- INB - Institut für Nano- und Biotechnologien (5)
- IfB - Institut für Bioengineering (3)
- Fachbereich Chemie und Biotechnologie (2)
- Fachbereich Elektrotechnik und Informationstechnik (2)
- Fachbereich Luft- und Raumfahrttechnik (2)
- ECSM European Center for Sustainable Mobility (1)
- Institut fuer Angewandte Polymerchemie (1)
The growing body of political texts opens up new opportunities for rich insights into political dynamics and ideologies but also increases the workload for manual analysis. Automated speaker attribution, which detects who said what to whom in a speech event and is closely related to semantic role labeling, is an important processing step for computational text analysis. We study the potential of the large language model family Llama 2 to automate speaker attribution in German parliamentary debates from 2017-2021. We fine-tune Llama 2 with QLoRA, an efficient training strategy, and observe our approach to achieve competitive performance in the GermEval 2023 Shared Task On Speaker Attribution in German News Articles and Parliamentary Debates. Our results shed light on the capabilities of large language models in automating speaker attribution, revealing a promising avenue for computational analysis of political discourse and the development of semantic role labeling systems.
Drought and water shortage are serious problems in many arid and semi-arid regions. This problem is getting worse and even continues in temperate climatic regions due to climate change. To address this problem, the use of biodegradable hydrogels is increasingly important for the application as water-retaining additives in soil. Furthermore, efficient (micro-)nutrient supply can be provided by the use of tailored hydrogels. Biodegradable polyaspartic acid (PASP) hydrogels with different available (1,6-hexamethylene diamine (HMD) and L-lysine (LYS)) and newly developed crosslinkers based on diesters of glycine (GLY) and (di-)ethylene glycol (DEG and EG, respectively) were synthesized and characterized using Fourier transform infrared (FTIR) spectroscopy and scanning electron microscopy (SEM) and regarding their swelling properties (kinetic, absorbency under load (AUL)) as well as biodegradability of PASP hydrogel. Copper (II) and zinc (II), respectively, were loaded as micronutrients in two different approaches: in situ with crosslinking and subsequent loading of prepared hydrogels. The results showed successful syntheses of di-glycine-ester-based crosslinkers. Hydrogels with good water-absorbing properties were formed. Moreover, the developed crosslinking agents in combination with the specific reaction conditions resulted in higher water absorbency with increased crosslinker content used in synthesis (10% vs. 20%). The prepared hydrogels are candidates for water-storing soil additives due to the biodegradability of PASP, which is shown in an exemple. The incorporation of Cu(II) and Zn(II) ions can provide these micronutrients for plant growth.