Overview and Problem Statement
The justices of the Supreme Court of the United States make decisions on cases that have a high societal impact every year. Transcripts of their oral argumentations present, therefore, an interesting resource to study argumentation patterns and question answering behaviors. The aim of this project is to develop a visual, semi-automatic learning method for the analysis of linguistic patterns of interest.
- Identification of linguistic patterns (questions, arguments, etc.) for detection.
- Extraction of features describing these patterns from the texts.
- Application of machine learning to generate a model for pattern classification/prediction.
- Development of a visual analytics interface to show findings.
- Good knowledge about information visualization and natural language processing.
- Experience in machine learning.
- Scope: Master
- 6 Month Project, 6 Month Thesis (Master)
- Start: immediately