GRAPHICAL VISUALIZATION OF RHETORICAL STRUCTURES IN CHATS
GRAPHICAL VISUALIZATION OF RHETORICAL STRUCTURES IN CHATS

Summary/Abstract: Chat Conversations are now used for different purposes including Computer-Supported Collaborative Learning (CSCL). In their analysis it is very important to identify the discourse threads and the rhetorical structures. In this paper we present an application for graphical visualization of rhetorical structures in chats. This software tool provides structured visualization and analysis of Computer-Supported Collaborative Learning chats. In our application, we first use the implemented tools that are based on the ideas of Mikhail Bakhtin’s dialog theory and Ştefan Trăuşan-Matu’s polyphonic model in order to analyze the chats and identify their topics and discussion threads (repeated words). In addition we identify cue phrases that are used in order to detect rhetorical schemata. As a result we determine the rhetorical relations for each one of the identified cue phrases (Conclusion, Contrast, Extending, Sequence, Explanation, etc.), so that the graphics are representing the links between the threads and the cue phrases. The resulting links are differentiated by colors for easing to detect the rhetorical relation between threads. Furthermore we visualize the links between threads with cue phrases that are important for CSCL in order to analyze the involvement of students in the chat, as well as their ability to track threads into the conversation and the extent the coherence among themselves. Additionally, we shall study graphics from the application in order to generate feedback for students. We will notice that the application analyzes various chats from the K-Teams laboratory of University Politehnica Bucharest.

• Details
• Contents