A5061659312- AI in Service Interactions
- Topic Modeling
- Digital Mental Health Interventions
- Natural Language Processing Techniques
- Maritime Navigation and Safety
- Speech and dialogue systems
- Mental Health via Writing
- Mobile Health and mHealth Applications
- Sentiment Analysis and Opinion Mining
- Text Readability and Simplification
- Mobile Learning in Education
- Narrative Theory and Analysis
- Visual and Cognitive Learning Processes
- Diet and metabolism studies
- Maritime Security and History
- Diabetes Treatment and Management
- Mental Health Research Topics
- Regulation of Appetite and Obesity
- Breastfeeding Practices and Influences
- Health Literacy and Information Accessibility
- Recommender Systems and Techniques
- Innovative Human-Technology Interaction
- Emotion and Mood Recognition
- Media, Gender, and Advertising
- Persona Design and Applications
University of Regensburg
2021-2024
Fraunhofer Institute for Experimental Software Engineering
2018
Scarcity of user data continues to be a problem in research on conversational interfaces and often hinders or slows down technical innovation. In the past, different ways synthetically generating data, such as augmentation techniques have been explored. With rise ever improving pre-trained language models, we ask if can go beyond methods by simply providing appropriate prompts these general purpose models generate data. We explore feasibility cost-benefit trade-offs using non fine-tuned...
The Future Conversations workshop at CHIIR'21 looked to the future of search, recommendation, and information interaction ask: where are opportunities for conversational interactions? What do we need get there? Furthermore, who stands benefit? was hands-on interactive. Rather than a series technical talks, solicited position statements on opportunities, problems, solutions in search all modalities (written, spoken, or multimodal). This paper -co-authored by organisers participants workshop-...
Valuable insights into an individual's current thoughts and stance regarding behaviour change can be obtained by analysing the language they use, which conceptualized using Motivational Interviewing concepts. Training conversational agents (CAs) to detect employ these concepts could help them provide more personalized effective assistance. This study investigates similarity of written around spanning diverse social contexts objectives. Drawing on previous research that applied MI texts about...
Conversational agents are increasingly used to address emotional needs on top of information needs. One use case increasing interest counselling-style mental health and behaviour change interventions, with large language model (LLM)-based approaches becoming more popular. Research in this context so far has been largely system-focused, foregoing the aspect user impact can have LLM-generated texts. To issue, we share a dataset containing text-based interactions related two GPT-4-based...
Conversational agents are increasingly used to address emotional needs on top of information needs. One use case increasing interest counselling-style mental health and behaviour change interventions, with large language model (LLM)-based approaches becoming more popular. Research in this context so far has been largely system-focused, foregoing the aspect user impact can have LLM-generated texts. To issue, we share a dataset containing text-based interactions related two GPT-4-based...
Motivational factors play a significant role in weight loss. Providing appropriate support based on an individual's motivational level could be deciding factor loss success. Considering the high cost of obesity to public health care, system with ability assess motivation benefit individuals and alike. As self-report measures are not always available would tedious use regular basis, one way achieve this understanding interpretation natural language. This project studies feasibility using...
This article examines the quality of weight loss information on German language web and studies how websites, likely to be accessed via popular search engines, are evaluated by end users. Sixty-five websites were identified qualitatively examined with respect content as defined literature, well meta design structure page. In a further step, same pages non-expert users in an online study. Deficiencies found, both terms behaviour rating competence Many showed little or no relevance for 46%...
In this demo, we present an interactive recommender system that suggests recipes to participants of a weight loss programme. Nutritional constraints imposed by the programme serve as initial information tackle crucial cold start problem.
Success of weight loss programmes can be highly dependent on the provision appropriate information. The guidance individuals should receive varies depending their motivational state. If information systems were able to discern a user's current stage change and use this retrieve support strategies, could greatly increase impact implementation maintenance behavioural changes necessary for successful loss. This PhD project explores feasibility predicting person's in context based personal they...
Refugees trying to get Europe via the Mediterranean often face human rights violations. The present situation is not in line with UN's SDG's 10 and 16. We Argo: a semi-automatically created vessel classification dataset focused on small boats, aim enable NGOs public detect refugee boats satellite imagery. achieve recall of 91% ships. With tool developed top results presented here, could collect information hold institutions participating illegal activities accountable.
In this description paper we outline the system architecture submitted to Task 4, Subtask 1 at SemEval-2022. We leverage generative power of state art pretrained transformer models increase training set size and remedy class imbalance issues. Our best is trained on a synthetically enhanced dataset with 10.3 times as many positive samples original reaches an F1 score 50.62%, which 10 percentage points higher than our initial undersampled version dataset. explore possible reasons for...
The First Early Career Researchers Roundtable for Information Access Research Workshop , in conjunction with the Seventh ACM Conference on Human Interaction and Retrieval (CHIIR) 2022, looked into future of research, collaborations, self-development to ask following. Where are opportunities researchers a (post-)pandemic environment, especially (ECRs)? What do we need get there? Which practical implementations can broader CHIIR community support? workshop started an invited talk. Instead...