Diplomarbeit DIP-2973

Kessler, Wiltrud: Analysis and Visualization of Coreference Features.
Universität Stuttgart, Fakultät Informatik, Elektrotechnik und Informationstechnik, Diplomarbeit Nr. 2973 (2010).
75 Seiten, englisch.
CR-Klassif.I.2.7 (Natural Language Processing)
I.5.1 (Pattern Recognition Models)
I.5.4 (Pattern Recognition Applications)

In this work, a new visualization for coreference has been presented and implemented in Coalda. This visualization allows the user to interactively label links with coreference information and to explore the space of the features used to create the visualization. The visualization works on feature vectors. To create these feature vectors, markables are extracted from a text in natural language. These markables are paired to form links. For every link, a feature vector is calculated based on a feature set. This preprocessing is done by modules of the SÜKRE project. In Coalda, the resulting feature vectors are then mapped to a SOM and this SOM is displayed to the user. The SOM is visualized as a graph. The nodes of the SOM are nodes of the graph and the neighbourhood relations in the output space are the edges. Distances between nodes are visualized by using the U-matrix value of nodes and edges to color them.

Volltext und
andere Links
PDF (1083767 Bytes)
Zugriff auf studentische Arbeiten aufgrund vorherrschender Datenschutzbestimmungen nur innerhalb der Fakultät möglich
Abteilung(en)Universität Stuttgart, Institut für Visualisierung und Interaktive Systeme, Visualisierung und Interaktive Systeme
BetreuerBurkovski, Andre
Eingabedatum27. April 2010
   Publ. Informatik