This chapter from a textbook of search user interfaces discusses techniques, ways for visualizing text or documents. Visualizing text or documents will definitely help visualize important parts of the documents, key concepts, etc. Though I agree with the author that mixing up data analysis objectives with text visualization will result in bad tools and design. The TAKMI interface seems to be very organized, structured and allowing users to select and choose how they want to see the information. The chapter discusses other tools like TRIST, BETA, of which I find TRIST very user friendly. It seems like a tool which effectively brings out the major entities and relationships between them, while continuing to provide the users some degree of flexibility as to how they want to view or filter the information.
The next section includes examples for visualizing word frequencies and concordances. Many were discussed in class as visualization examples. Of all, I find the DocuBurst visualization interesting as it allows us to see the context of the word as well. Classic theme river visualization - the name voyager for baby names is also discussed. Lastly, the author discusses how citations and their relationships can be visualized. This seems to be more complicated and requiring precise outputs.
In all, I feel visualization for text and document analysis is a very interesting application domain for information visualization. It could help in recognition of patters, frequencies, relationships with greater precision and more systematically than just opportunistic discoveries.
The next section includes examples for visualizing word frequencies and concordances. Many were discussed in class as visualization examples. Of all, I find the DocuBurst visualization interesting as it allows us to see the context of the word as well. Classic theme river visualization - the name voyager for baby names is also discussed. Lastly, the author discusses how citations and their relationships can be visualized. This seems to be more complicated and requiring precise outputs.
In all, I feel visualization for text and document analysis is a very interesting application domain for information visualization. It could help in recognition of patters, frequencies, relationships with greater precision and more systematically than just opportunistic discoveries.
0 comments:
Post a Comment