Raheleh Makki Niri
Dr. Evangelos E. Milios
Dr. Stephen Brooks
Traditional document clustering is an unsupervised task which tries to group similar documents of a given collection in the way that documents within the same category are more similar than those in other categories. These techniques do not use any user interaction; consequently, it is expected that the generated clusters may not always match with the user’s conception of the collection.
I am currently working on utilizing visualization in text document clustering. Showing the obtained clusters to the user and involving her in the clustering process is the goal. The user can interact with the visualization and make the results closer to her expectation.