Dynamic UI Element Generation for Rapid Prompting
bachelor thesis
Status | in progress |
Student | Oliver Hans Kühle |
Advisor | Rifat Amin |
Professor | Prof. Dr. Andreas Butz |
Task
Problem Statement
This thesis aims to facilitate prompting through dynamic user interface (UI) element generation, aimed at enhancing user experience and productivity in generating diverse textual content. Traditional prompt-based creative writing tools often provide static prompts, limiting user flexibility and creativity. This thesis will addresses this limitation by dynamically generating UI elements based on actionable words identified within the prompt. These actionable words encompass parameters such as text type (e.g., story, poem), word count, language, and other relevant attributes. Leveraging natural language processing techniques, the developed system will parse the prompt to identify these actionable elements, which are then used to generate a customized UI for the user. Through this adaptive UI, users can easily modify and experiment with various aspects of the prompt, such as changing the protagonist's characteristics or altering the narrative structure, fostering creativity and exploration. We discuss the design and implementation of our system, highlighting its flexibility and ease of use. Furthermore, we evaluate the effectiveness of our approach through user studies, demonstrating its capability to empower writers in generating rich and diverse textual content. Overall, our work contributes to the advancement of interactive creative writing tools, opening up new avenues for personalized and immersive writing experiences.
Tasks
- Perform a literature review. Identify key challenges and gaps in current approaches to prompt-based creative writing tools
- Define algorithms and data structures for parsing prompts, identifying actionable words, and generating corresponding UI elements
- Implement an interactive website for the users
- Implement outlier detection and clustering algorithms in order to identify interesting information from the data
- Conduct a user study
- Do a statistical evaluation
- Write a thesis and present your findings in the Disputationsseminar
- (Optional:) co-write a research paper